US20190317613A1 - Apparatus for controlling contents of a computer-generated image using 3d measurements - Google Patents

Abstract

A method for controlling a computer display cursor in an interaction region includes projecting an image of the computer display to create the interaction region. A distance is established between a first point and a second point. The first point has a predetermined relation to the projection device, and the second point has a predetermined relation to the interaction region. At least one of an orientation and a position of a pointing line is measured. The pointing line has a predetermined relation to a pointing device. The established distance and the at least one of measured position and orientation are used to control the cursor position on the interaction region.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS
• This application is a continuation of U.S. application Ser. No. 15/014,879, filed Feb. 3, 2016, which is a continuation of U.S. application Ser. No. 11/483,170 filed Jul. 7, 2006, now U.S. Pat. No. 9,285,897, which claims priority to U.S. Provisional Application No. 60/698,702 filed Jul. 13, 2005, and U.S. Provisional Application No. 60/742,393 filed Dec. 5, 2005, all of which are hereby incorporated by reference.
STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT
• Not applicable.
BACKGROUND OF THEINVENTION1. Field of the Invention
• The invention relates generally to devices for making presentations in front of audiences and, more specifically, to devices and methods for making presentations for which interaction with the displayed information. through direct-pointing is desired.
2. Background Art
• Technology for presenting computer-generated images on large screens has developed to the point where such presentations are commonplace. For example, the software package POWERPOINT, sold by Microsoft Corp., Redmond, Wash., may be used in combination with a so-called ‘beamer’ to generate interactive presentation material and project for viewing by an audience. Often, such presentations are held in rooms not specifically equipped for the purpose of showing presentations, in which case use is made of a portable beamer in combination with a so-called ‘laptop’ computer. Under these circumstances the projection surface may be a wall of the room.
• During a presentation it is desirable for the presenter to be able to move freely in front of the audience while retaining the capability to interact with the presentation and point to specific features on the displayed images. In general, interaction with a computer is often facilitated by pointing devices such as a ‘mouse’ or a ‘trackball’ that enable manipulation of a so-called ‘cursor’. Traditionally, these devices were physically connected to the computer, thus constraining the freedom-of-movement of the user. More recently, however, this constraint has been removed by the introduction of wireless pointing devices such as the GYROMOUSE, as manufactured by Gyration, Inc.
• Broadly speaking, pointing devices may be classified in two categories: a) devices for so-called ‘direct-pointing’ and b) devices for so-called ‘indirect-pointing’. Direct pointing devices are those for which the physical point-of-aim coincides with the item being pointed at, i.e., it lies on the line-of-sight. Direct pointing devices include the so-called ‘laser pointer’ and the human pointing finger. Indirect pointing devices include systems where the object of pointing (e.g., a cursor) bears an indirect relationship to the physical point-of-aim of the pointing device; examples include a mouse and a trackball. It needs no argument that direct-pointing systems are more natural to humans, allowing faster and more accurate pointing actions.
• Indirect pointing devices known in the art include the following. U.S. Pat. No. 4,654,648 to Herrington et al. (1987), U.S. Pat. No. 5,339,095 to Redford (1994), U.S. Pat. No. 5,359,348 to Pilcher et al. (1994), U.S. Pat. No. 5,469,193 to Giobbi et al. (1995), U.S. Pat. No. 5,506,605 to Paley (1996), U.S. Pat. No. 5,638,092 to Eng et al. (1997), U.S. Pat. No. 5,734,371 to Kaplan (1998), U.S. Pat. No. 5,883,616 to Koizumi et al. (1999), U.S. Pat. No. 5,898,421 to Quinn (1999), U.S. Pat. No. 5,963,134 to Bowers et al. (1999), U.S. Pat. No. 5,999,167 to Marsh et al. (1999), U.S. Pat. No. 6,069,594 to Barnes et al. (2000), U.S. Pat. No. 6,130,664 to Suzuki (2000), U.S. Pat. No. 6,271,831 to Escobosa et al. (2001), U.S. Pat. No. 6,342,878 to Chevassus et al. (2002), U.S. Pat. No. 6,388,656 to Chae (2002), U.S. Pat. No. 6,411,277 to Shah-Nazaroff (2002), U.S. Pat. No. 6,492,981 Stork et al. (2002), U.S. Pat. No. 6,504,526 to Mauritz (2003), U.S. Pat. No. 6,545,664 to Kim (2003), U.S. Pat. No. 6,567,071 to Curran et al. (2003) and U.S. Patent Application Publication No. 2002/0085097 to Colmenarez et al. (2002). Each of the foregoing publications discloses a system for which the 2 dimensional or 3 dimensional position orientation and/or motion of an object, such as a handheld pointing device, are measured with respect to some reference coordinate system using appropriate means. Such means include acoustic devices, electromagnetic devices, infrared devices, visible light emitting diode (LED) devices, charge coupled devices (CCD), accelerometer and gyroscopic motion detectors, etc. Although for some of the foregoing devices the reference coordinate system may be positioned close to the display means, no information on the actual position of the presentation display with respect to the system is used, causing the resulting pointing action to be inherently indirect and, hence, less natural to the human operators of such systems.
• Other inherently indirect-pointing systems that do not require the position or orientation of the pointing device to be known include devices such as disclosed in U.S. Pat. No. 5,095,302 to McLean et al. (1992) and U.S. Pat. No. 5,668,574 to Jarlance-Huang (1997). The foregoing patents describe indirect-pointing methods that do not provide the speed and intuitiveness afforded by direct-pointing systems.
• Direct pointing devices are disclosed, for example, in U.S. Pat. No. 4,823,170 to Hansen (1989), which describes a direct-pointing system comprising a light source, a position-sensing detector located adjacent to the light source and a focusing reflector that, in one application, is parabolic shaped and is attached to the pointing device. Additionally, procedures are described to calibrate the system. In the understanding of current applicant, however, the physical location of the position-sensing detector needs to be, at least preferably, adjacent to the display means. The system disclosed in the Hansel '170 patent cannot easily be ported to a room not specifically equipped for this system.
• U.S. Pat. No. 5,929,444 to Leichner (1999) discloses a system primarily intended for target shooting practice, but an application as a direct-pointing cursor control apparatus may arguably be anticipated. The system includes transmitting and detecting equipment in a fixed reference base and a moveable pointing device. A calibration routine is described that aligns the detecting and transmitting means while keeping the pointing device (i.e., a gun) aimed at the center of the target. The Leichner '444 patent does not describe methods or means that allow determination of a point-of-aim of a pointing device on a target of which the size and/or orientation have not been predetermined. Consequently, the system disclosed in the Leichner '444 patent is not suitable to be used as a cursor control means for projection surfaces not specifically adapted to be used with such a system.
• U.S. Pat. No. 5,952,996 to Kim et al. (1999), U.S. Pat. No. 6,184,863 to Sibert et al. (2001) and US Patent Application Publication No. 2002/0084980 to White et al. (2002) disclose direct-pointing systems where the 3 dimensional position and/or orientation of the pointing device is measured with respect to sources and/or detectors, the physical position of which in relation to the display means is presumed known. Such systems only work in rooms specifically equipped for their use.
• U.S. Pat. No. 5,484,966 to Segen (1996), U.S. Pat. No. 6,335,723 to Wood et al. (2002) and U.S. Pat. No. 6,507,339 to Tanaka (2003) disclose methods suitable for direct-pointing that are useful only if the pointing device is physically close to or touching the display area or volume, for example used with so-called ‘interactive whiteboards’. Some of the foregoing patents describe appropriate pointer calibration routines. Such systems are not suitable for presentations where the display surface is out of the presenter's physical reach.
• U.S. Pat. No. 6,104,380 to Stork et al. (2000) discloses a direct-pointing system in which at least the 3 dimensional orientation of a handheld pointing device is measured by appropriate means. Additionally, a direct measurement is made of the distance between the pointing device and the displayed image. However, the system disclosed in the Stork et al. '380 patent does not include methods to ascertain the position and orientation of the display means relative to the pointing device. In the foregoing system, these appear to be presumed known. This is also the case for a system disclosed in U.S. Pat. No. 4,768,028 to Blackie (1988), in which the orientation of a helmet-mounted direct-pointing apparatus is measured electromagnetically. The foregoing systems therefore appear to be ill-suited to operate in rooms not specifically equipped for presentation purposes.
• U.S. Pat. No. 6,373,961 to Richardson et al. (2002) discloses a direct-pointing system using helmet-mounted eye tracking means. The point-of-gaze relative to the helmet is measured as well as the position and orientation of the helmet relative to the display. The latter is accomplished by equipping the helmet either with means to image sources mounted at fixed positions around the display or with means to image a displayed calibration pattern. Of course, the foregoing system relies on sophisticated helmet mounted equipment capable of, among other things, tracking eye-movement. Moreover, such a system relies on an unobstructed line-of-sight with respect to the display and a substantially constant distance from the display to the helmet-mounted equipment. The disclosed invention does not lend itself to be easily used by a human operator in an arbitrary (not predetermined) presentation setting.
• U.S. Pat. No. 6,385,331 to Harakawa et al. (2002) discloses a system that uses infrared technology in combination with image recognition software to distinguish pointing gestures made by the presenter, without the need for an artificial pointing device. The disclosed system, however, requires the presentation room to be set up with highly tuned and sophisticated equipment, and is therefore not easily ported to a different venue.
• U.S. Pat. No. 6,404,416 to Kahn et al. (2002) discloses a direct-pointing system where a handheld pointing device is equipped with an optical sensor. In such system either the display is required to be of a specific type (e.g., a cathode ray-based display that uses an electron: beam) or the displayed image is required to be enhanced by timed and specific emanations. When pointed to the display, a handheld pointing device may detect the electron beam or the timed emanations, and the timing of these detections may then be used to ascertain the point-of-aim. The disclosed system is somewhat similar to the technologies used for so-called light guns in video games as disclosed, for example, in U.S. Pat. No. 6,171,190 to Thanasack et al. (2001) and U.S. Pat. No. 6,545,661 to Goschy et al. (2003). Of course, such systems require either a specific display apparatus or a specialized modification of the displayed images. Moreover, an uncompromised line-of-sight between the pointer and the display is a prerequisite for such systems.
• U.S. Patent Application Publication No. 2002/0089489 to Carpenter (2002) discloses a direct-pointing system that, in one embodiment, positions a computer cursor at a light spot projected by a laser-pointer. The system relies on the use of an image-capturing device to compare a captured image with a projected image. As such, the system disclosed in the '489 application publication makes use of calibration routines in which the user is required to highlight computer-generated calibration marks with the laser pointer. The system disclosed in the '489 patent application publication is not unlike a system disclosed in U.S. Pat. No. 5,502,459 to Marshall et al. (1996). Also, U.S. Pat. No. 5,654,741 to Sampsell et al. (1997), U.S. Pat. No. 6,292,171 to Fu et al. (2001), U.S. Patent Application Publication No. 2002/0042699 to Tanaka et al. (2002) and U.S. Patent Application Publication No. 2002/0075386 to Tanaka (2002) all disclose systems that can detect a light-spot using optical means. Such systems specifically generally require the use of computationally expensive image processing technologies. All of these inventions require a projection surface with adequate diffusion properties, as well as some form of optical system with a steady and uncompromised view of the display area. As such, they limit the freedom-of-movement of the presenter and place limitations on the position and optical characteristics of the necessary equipment. Also, in some of these inventions fast and involuntary movement of the user's hand may result in a cursor that does not move smoothly or a cursor that does not perfectly track the light spot, causing possible confusion with the user.
• U.S. Patent Application Publication No. 2005/0270494 to Banning (2005) and pages 83-87 of ISSN 1476-2986 by Ashdown (2004) describe direct-pointing methods for which the spatial characteristics of a projected display are determined by means of a user-executed calibration procedure. Direct-pointing cursor control is subsequently realized by measuring the position and/or orientation of a pointing device and mathematically intersecting the associated pointing line with the projected display. The calibration procedure requires the user to direct a sufficient number of pointing lines through characteristic points of the projected display, after which its spatial properties are calculated geometrically. To this end, Ashdown's method requires the user to register the spatial properties of 9 lines, whereas Banning uses substantially less lines in combination with assumptions on shape, orientation and/or position of the projected display. Both methods, however, rely on either measuring the position of the pointing device or on the requirement that this position remains constant. As such, they either rely on sophisticated and expensive tracking equipment or severely constrain the freedom-of-movement of the user.
• Other pointing systems known in the art may be classified as other than entirely direct-pointing or indirect-pointing systems. Such systems include one disclosed in U.S. Pat. No. 6,417,840 to Daniels (2002), which is combination of a cordless mouse with a laser pointer. Although this system incorporates a direct-pointing device (i.e., the laser pointer), the method used by the system for interacting with the presentation is indirect (i.e., by means of the mouse) and therefore does not afford the fast and more accurate interactive pointing actions provided by some other direct-pointing systems described in some of the foregoing publications.
• Another system known in the art that uses both direct and indirect-pointing methods is described in U.S. Pat. No. 6,297,804 to Kashitani (2001). The disclosed system is a system for pointing to real and virtual objects using the same pointing device. In the disclosed system, the pointing means switch between a computer controlled light source (e.g., a laser) and a conventional computer cursor, depending on whether or not the user's intended point-of-aim has crossed a boundary between the real and virtual display (i.e., computer-displayed imagery). Various methods are described to establish these boundaries. Although the computer-controlled light source may be regarded as pointing in a direct manner, its point-of-aim is essentially governed by the system operator using an indirect-pointing system such as a mouse. Thus, the disclosed system does not allow for the desired flexibility afforded by truly direct-pointing methods.
• Other systems known in the art include those such as disclosed in U.S. Pat. No. 5,796,386 to Lipscomb et al. (1998), which discloses a calibration procedure to establish the relation between coordinate systems associated with a handheld device and, in one embodiment, a display unit. The system disclosed in the Lipscomb et al. '386 patent may arguably be applicable as a direct-pointing cursor control system. The disclosed calibration routine requires the user to register at least three 3 dimensional positions of the handheld device with a central processing unit. The disclosed system does not appear to include calibration methods for the case where the display unit is out of physical reach of the system operator. The system is thus not practical for use at an arbitrary venue.
• U.S. Pat. No. 6,084,556 to Zwem (2000) discloses a head-mounted display unit that displays information governed by the orientation of the apparatus, as measured by a tracking device. This way, the system creates the illusion of a large virtual display that is being looked at by the system operator. Of course, the large display alluded to does not constitute a projected presentation. Also, no methods are disclosed in the Zwem '556 patent to establish the relative position of the head-mounted apparatus with respect to the display.
• U.S. Patent Application Publication No. 2002/0079143 to Silverstein et al. (2002) discloses a system in which the position of a moveable display with respect to a physical document is established. The '143 patent application publication describes calibration routines in which the user is required to activate a registration button when the moveable display is over some predetermined position on the physical document. The disclosed system only relates to 2 dimensional applications and, moreover, cannot be used in situations where the interaction region is out of the system operator's physical reach.
SUMMARY OF THE INVENTION
• One aspect of the invention is a method for controlling a computer display cursor in an interaction region. A method according to this aspect of the invention includes projecting an image of the computer display to create the interaction region. A distance is established between a first point and a second point. The first point has a predetermined relation to the projection device, and the second point has a predetermined relation to the interaction region. At least one of an orientation and a position of a pointing line is measured. The pointing line has a predetermined relation to a pointing device. The established distance and the at least one of measured position and orientation are used to control the cursor position on the interaction region.
• Another aspect of the invention is a method for controlling a parameter related to the position of a cursor on a projected computer screen image. According to this aspect, the method includes establishing a first point having a predetermined relation to the projected computer screen image. An orientation of a pointing line is measured while directing the pointing line to substantially pass through the first point. The pointing line has a predetermined relation to a pointing device. A positional coordinate of the pointing device, relative to the projected computer screen image, is constrained using the measurement of the orientation of the pointing line, and the positional coordinate of the pointing device is used to control the parameter of said cursor.
• Another aspect of the invention is a method for controlling a computer display cursor in an interaction region. The method according to this aspect includes establishing a first and a second point. Both the first and the second point have a predetermined relation to the interaction region. A pointing line with a predetermined relation to a pointing device is directed to substantially pass through the first point, and its orientation is measured. The measurement is used to constrain a positional coordinate of the pointing device. The orientation of the pointing line is also measured while directing it to substantially pass through the second point, and this measurement is also used to constrain a positional coordinate of the pointing device. No more than two positional coordinates of the pointing device are initialized to have a predetermined relation during the first and the second measurement of the orientation of the pointing line. The positional coordinates of the pointing device are used. to control the computer display cursor.
• Other aspects and advantages of the invention will be apparent from the following description and the appended claims.
BRIEF DESCRIPTION OF THE DRAWINGS
• FIG. 1 shows a pointing device and base station according to a first embodiment.
• FIG. 2 shows a presentation venue and computer screen.
• FIG. 3 shows program steps for selection of appropriate assumptions according to several embodiments.
• FIG. 4 shows program steps for construction of information set according to several embodiments.
• FIG. 5 shows program steps for ascertaining necessary 3D data according to several embodiments.
• FIG. 6 shows program steps for control of pointing device elements and computer cursor according to several embodiments.
• FIG. 7 shows one example of a first embodiment.
• FIG. 8 shows a second example of one embodiment of the invention.
• FIG. 9 shows a third example of one embodiment.
• FIG. 10 shows a pointing device and base station according to a second embodiment.
• FIG. 11 shows program steps for construction of information set according to several embodiments.
• FIG. 12 shows program steps for ascertaining necessary 3D data according to several embodiments.
• FIG. 13 shows a pointing device and base station according to a third embodiment.
• FIG. 14 shows program steps for ascertaining necessary 3D data according to several embodiments.
• FIG. 15 shows projection of interaction structure on horizontal and vertical planes.
• FIG. 16 shows an image of calibration point and light spots at the points-of-aim.
• FIG. 17 shows an alternative embodiment of pointing device.
• FIG. 18 shows a prior art: construction of method M for a quadrangle.
• FIG. 19 shows a fourth preferred embodiment.
• FIG. 20 shows prior art: the correction of vertical and horizontal keystoning.
• FIG. 21 shows application of the cosine rule for non-perpendicular projection situations.
• FIG. 22 shows a pointing device without base station according to a fifth embodiment.
• FIG. 23 shows a fifth preferred embodiment.
• FIG. 24 shows a sixth preferred embodiment.
• FIG. 25 shows a seventh preferred embodiment.
DETAILED DESCRIPTION
• A first embodiment of the invention will be described with reference toFIG. 1. Apointing device20 has associated with it a coordinate system x′ y′ z′. Aportable base station30 has associated with it a coordinate system x y z. Pointingdevice20 andbase station30 may be equipped with coordinate sensing devices,201 and301, respectively, that enable measurement of the 3 dimensional position and 3 dimensional orientation of pointingdevice20 and, therefore, of a pointing line21 (see alsoFIG. 2) that substantially intersects pointingdevice20, all measured relative to the x y z coordinate system. Pointingline21 may substantially coincide with the long axis of pointingdevice20. For example, coordinatesensing devices201 and301 may be electromagnetic tracking devices, such as the 3SPACE FASTRAK® system, manufactured by Polhemus, a Kaiser Aerospace & Electronics Company, Colchester, Vt. Alternatively, coordinatesensing device201 and301 may be based on (ultra)sonic tracking systems such as those used in the LOGITECH 2D/6D computer mouse, commercially available from Logitech Inc., Fremont, Calif. Other embodiments for coordinatesensing device201 and30 I may include, without limitation, sensors based on LEDs, CCDs, accelerometers; inclinometers, gyroscopes, compasses, magnetometers, etc. Also, combinations of such different types of sensors may be used to acquire redundant measurements for quality control purposes. For example, a controlled-source electromagnetic position tracking system may be combined with compasses and inclinometers. Those skilled in the art will appreciate that independent measurements of the Earth's magnetic field and gravity may be used to enhance the measurements made by a controlled-source position detecting system. In the invention, any system may be used that is capable of determining at least parts of the orientation and position in three dimensions, with respect to coordinate system x y z, of a line-segment that substantially intersects pointingdevice20.
• Base station30 may comprise a plurality of related physical entities, such as described in U.S. Pat. No. 6,608,668 to Faul et al (2003); for clarity of explanation one of these physical entities will be associated with coordinate system x y z and be denoted asbase station30, the center of which substantially coincides with the origin of coordinate system x y z. For purposes of explanation of the invention, the origin of coordinate system x′ y′ z′ substantially coincides with the center of pointingdevice20, and the z′-axis is substantially aligned with the long axis of pointingdevice20.
• Pointingdevice20 may also be provided with a light-beam projector202, for example a laser. The physical position and orientation of the projected beam of light with respect to coordinate system x′ y′ z′ may be established with suitable accuracy at the place of manufacture of thepointing device20 and may be presumed to be known. For purposes of explanation, the beam of light from thelight beam projector202 substantially coincides with the z′-axis. Additionally, one ormore control buttons203a,203bor the like may be provided, as well as communication andcontrol device204. Communication andcontrol device204 may be used to control various features of pointingdevice20 and may also communicate via wire, or wirelessly, withbase station30, a central processing unit (not shown separately) such as a COMPAQ Armada M700 as manufactured by Hewlett Packard Company, Palo Alto, Calif., or any other components of the system. The central processing unit may also control a presentation with which user interaction is desired. For clarity of the description which follows, the central processing unit will be referred to hereinafter as “the computer.” Furthermore, pointingdevice20 may include aninternal power source205, for example a battery, or instead may be connected such as by wire to an external power source.
• In addition to coordinatesensing device301,base station30 may be provided with communication andcontrol device304. Communication andcontrol device304 may be used to control various features ofbase station30. Communication andcontrol device304 may communicate with pointingdevice20, with the computer (not shown) that may control the presentation images, or with any other components of the system. The communication andcontrol device304 may use wire- or wireless technology. In the present embodiment communication with the computer occurs via a universal serial bus (USB) compatible device or the like (not shown). Communication andcontrol device304 may communicate wirelessly with a relay device (not shown) that communicates via a USB compatible device or the like with the computer (not shown) that may control the presentation with which user-interaction is desired. There may bevisible markings302 onbase station30 that indicate the orientation of one or more coordinate planes defined by prescribed relations between coordinates x, y, z, as well as features of the position of the origin of the x y z coordinate system. For example, a line may be used to indicate the z-x-plane of the x y z coordinate system, for which the y-coordinate is zero; another line may be used to indicate the z-y-plane, for which the x-coordinate is zero.Base station30 may be provided with a level-sensing device303 to determine whether or not one of the coordinate planes of coordinate system x y z is substantially horizontal. Level-sensing device303 may measure two angles or equivalent characteristics that define the orientation of one of the coordinate planes of coordinate system x y z with respect to a horizontal surface. Level-sensing device303 can be a device such as disclosed in U.S. Pat. No. 6,466,198 to Feinstein (2002), which makes use of model ADXL202 accelerometers sold by Analog Devices Inc., Norwood, Mass. The disclosed accelerometers provide tilt angle information depending on their inclination relative to Earth's gravity. It will be apparent to those skilled in the art that many other types of sensors may be used as embodiment for level-sensing device303, for example, capacitance-type bubble levels. See, for example, U.S. Pat. No. 5,606,124 to Doyle et al. Finally,base station30 may be equipped with apower source305, for example a battery, or may have a connection to an external power source.
• Referring toFIG. 2, aprojection device40 is arranged to project animage50 generated by, for example, the computer (not shown). Theprojection device40 may be used to generate aprojection image70 on aprojection region60. For example,projection device40 may be a 2000 lumen projector, model XL8U, manufactured by Mitsubishi Corp.Projection region60 may be a surface, such as a wall or a projection screen. Theprojection region60 may define a flat plane or a may define a more elaborate 2 dimensional (2D) or even 3 dimensional (3D) structure. InFIG. 2,projection region60 is shown as a screen, but this is only for purpose of illustrating the principle of the invention and is not intended to limit the scope of the invention. Alternatively the combination ofprojection device40 andprojection region60 may be incorporated in one and the same physical device, such as a television receiver (cathode ray tube display), liquid crystal display (LCD) screen or the like.
• There is a region of space that is designated as a region on which interaction with the user is desired. This region is denoted as theinteraction region71. Theinteraction region71 may be flat (planar) or may be a more elaborate 2D or even 3D structure. Theinteraction region71 may have features in common withprojection image70 and may be associated in some way with a computerscreen interaction region51. In the present embodiment, however, it will be assumed thatinteraction region71 or a scaled version thereof, is substantially part of or substantially coincides withprojection image70. Moreover, it will be assumed in this embodiment thatinteraction region71 substantially coincides with the projection of computerscreen interaction region51 as projected byprojection device40.
• For display systems which include aseparate projection device40 andprojection region60, the optical axis ofprojection device40 may not be aligned with any vector normal toprojection region60. Moreover,projection region60 may not be flat and14 therefore may not even be a 2D shape. Consequently,projection image70 andinteraction region71 may in general not be flat or rectangular, even if the imagery generated by the computer is scaled so as to be presented in rectangular form. In this embodiment, however, it is assumed thatprojection region60,projection image70 andinteraction region71 are substantially flat. Furthermore, it is assumed in this embodiment thatinteraction region71 is substantially a quadrangle and that the associated computerscreen interaction region51 is substantially rectangular.
• Additionally, calibration points721a,721b,721c,721dmay be provided that may define characteristic features ofinteraction region71. For example,interaction region71 may be trapezoidal in shape, in which case calibration points721a,721b,721c,721dmay define corners ofinteraction region71 or corners of a scaled version thereof. Furthermore, screen marks521a,521b,521c,521dmay be provided, and may but need not be associated with calibration points721a-721dFor example, calibration points721a-721dmay coincide with the projected versions of screen marks521a-521dand may in this way be identified byprojection device40. Calibration points721a-721dmay also be identified by other means than projection, for example by unique descriptions such as the ‘upper right corner of interaction region71’, ‘center of interaction region71’ etc.
• The operation of the present embodiment will now be described with reference toFIGS. 1, 2 and 3. A display system is set up at the venue where a presentation is to made, which can be a combination of aportable projection device40 andprojection surface60, for example a wall. The display system is connected, using appropriate interconnection devices, to the computer (which may be in thebase station30 or located elsewhere) that generates the presentation images.
• The system userpositions base station30 at a convenient location, preferably not far from where the user intends to conduct the presentation. The user may positionbase station30 in such a way that one of the coordinate planes of the x y z coordinate system is substantially parallel or substantially coincident withprojection region60. Thevisual markings302 may assist in such positioning. Subsequently, the user connectsbase station30 to the computer (not shown), for example via a USB compatible device connection (not shown), or using a wireless relay device (not shown). The computer may be disposed in thebase station30 in some embodiments. In some embodiments, the computer may recognize the base station connection and start a program, part of which may be contained in the communication andcontrol device304, in communication andcontrol device204, or in control logic contained in the wireless relay device (not shown). Alternatively, the user may be required to load the program into the computer manually via a CD-ROM drive, a floppy drive, memory stick (USB mass storage device not shown) or the like. However it is loaded into the computer, the program may initiate a calibration routine that has as its object establishment of the shape, position, size and orientation of a definedinteraction structure72 relative to the x y z coordinate system. Theinteraction structure72 is assumed to substantially coincide withinteraction region71. The operation of the program will now be explained with reference toFIG. 3.
• At80athe program is initiated. Duringstep80ba default assumption is made regardinginteraction region71. Specifically,interaction region71 is assumed to substantially coincide with a well-defined interaction structure72 (FIG. 2 shows aninteraction region71 and aninteraction structure72 that are clearly not identical; this difference is however only meant as an example, and as a practical matter is preferably as small as possible). At80bdefault values are entered for the orientation and position of thisinteraction structure72. For example, the default setting may assumeinteraction region71 to substantially coincide with aninteraction structure72 that is a (flat) quadrangle positioned in a vertical plane that substantially intersects the origin of the x y z coordinate system associated withbase station30. As another example, the default setting may provide thatinteraction region71 substantially coincides with aninteraction structure72 that is an isosceles trapezoid of which the parallel edges are horizontal and of which the position is unknown. Using such default values, calibration points721a-721dnot only define characteristic features ofinteraction region71 but also ofinteraction structure72.
• At80ca decision is made whether the default values forinteraction region71 andinteraction structure72 should be accepted or overridden by the user. In making this decision, input from level-sensing device303 andvisible markings302 onbase station30 may be used. If the defaults are to be accepted, the program continues to80j, the details of which are explained below with reference toFIG. 4. If the defaults are overridden, the program continues at80dto80i, during each of which the user may override any of the default settings. The user may be aided during any of80dto80iby a library of predetermined shapes, predetermined orientations and/or predetermined positions, or the user can be provided by the program with the capability to construct custom shapes, orientations and/or positions. In any case, the program continues to80j.
• It will be appreciated by those skilled in the art that once an assumption has been made regarding the shape ofinteraction structure72, it is possible to construct a set of three dimensionally distributed points in space that completely determines the 3D position, orientation and size of theinteraction structure72. The number of points in the set will depend on the complexity of the assumed shape and the ingenuity with which the points are chosen. For example, a rectangular shape that is arbitrarily oriented in space is completely determined by a set of 3 points that coincide with3 of its corners, but is also completely determined by a set of 8 points, pairs of which may determine each of the four edges of the rectangle.
• Referring toFIG. 4, describing details ofprogram element80j, the computer program continues at90a, at which a set P is generated that includes a quantity n of points in space, each represented as C(i) (0<i<n+1), that uniquely determineinteraction structure72, together with descriptions of their roles in determining this structure. For example, if the program elements outlined inFIG. 3 reveal thatinteraction region71 is assumed rectangular, set P may hold three points described as the upper-left corner, the lower-left corner and the upper-right corner of arectangular interaction structure72. If, alternatively, the projection of computerscreen interaction region51 is substantially rectangular, and this projected rectangle has the same center but is physically two times smaller thaninteraction structure72, the three points may be described as the upper-right corner, the lower-left corner and the lower-right corner of a rectangle that has the same center but is two times smaller thaninteraction structure72. Thus, by carefully choosing the points in set P together with their description, anyinteraction structure72 may be completely determined.
• In addition to set P, a 3D point CA and another 3D point CB are determined at90a. These additional 3D points are determined so as to lie away frominteraction region71, that is, they lie at some distance out of the plane that is closest to or substantially containsinteraction region71. The distance may be comparable to an edge ofprojection image70. For example, point CA may be determined to lie nearprojection device40 and point CB may be determined to lie at a distance from point CA that may be equal to the distance betweenprojection device40 andprojection surface60, measured substantially parallel toprojection surface60. Other choices for point CA and point CB may also be used. Additionally, sets A and B are generated duringstep90a. Set A includes a number n of lines A(i), each of which connects point CA to one of the points C(i) (0<i<n+1) in set P. Likewise, set B holds a number n of lines B(i), each of which connects point CB to one of the points C(i) (0<i<n+1) in set P. Finally, at90a, counters a, b are both initialized to the value of n.
• Flow then continues on to step90b, where a decision is made whether the program elements outlined inFIG. 3 have revealed any information on the 3D position and orientation ofinteraction structure72. If the decision is positive flow continues to step90c, where this information is used to establish a priori relationships between the coordinates of the points in set P. For example, the steps outlined inFIG. 3 may have revealed thatinteraction structure72 is assumed to coincide with the x-z plane, in which case the a priori relationships may include the requirement that the y-coordinates of all points in set P are equal to zero. A pre-set library of a priori relationships may be provided to aid in the execution of this program element.
• The program continues to90d, which may also be reached from90bif the decision at90bis negative. At90d, line B(b) is removed from set B, after which counter b is decremented by 1.
• The program continues to90e, where a decision is made whether complete 3D information on the lines in sets A and B, together with a priori information, constitutes enough information to uniquely determine the coordinates of all the points in set P. For example, the program elements outlined inFIG. 3 may have determined thatinteraction region71 is assumed to be rectangular, but no information is known regarding its orientation or position. In this case set P may contain three points, corresponding to three corners of arectangular interaction structure72. Each point C(i) (0<i<4) in set P would then be uniquely determined by the intersection of a line from set A and a line from set B if and only if sets A and B contained three lines each.
• If the decision at90eis negative, the program continues to90f, where line B(b+1) is added once more to set B.
• If the decision at90eis positive, program flow continues to90g. The program flow may also continue from90fto90g. At90ga decision is made whether counter b has reached zero, in which case the lines left in set B (which number may be equal to zero) are deemed necessary for unique determination of the coordinates of all the points in set P. If the decision at90gis positive, program flow continues to90h. If the decision at90gis negative, program flow reverts to90d.
• During90h, line A(a) is removed from set A, after which counter a is decremented by 1. Program flow then continues to90i, where a decision is made whether complete 3D information on the lines in sets A and B, together with the a priori information, constitute enough information to uniquely determine the coordinates of all the points in set P.
• If the decision at90iis negative, program flow continues to step90j, where line A(a+1) is added once more to set A.
• If the decision at90iis positive, program flow continues to step90k. The program flow also continues from90jto90k. At90k, a decision is made whether counter a has reached zero, in which case the lines left in set A (which number may be equal to zero) are deemed necessary for unique determination of the coordinates of all the points in set P. If the decision at90kis negative, program flow reverts to90h. If the decision at90kis positive, program flow continues to90m, the details of which are described with reference toFIG. 5.
• InFIG. 5, the program continues to100a, where counter p is initialized to 1 and variable n is set to the number of points in set P.
• Program flow then continues to100bwhere a decision is made whether line A(p) (connecting point CA to point C(p)) is included in set A. If the decision is negative, program flow continues to100c, where counter p is incremented by 1.
• If the decision at100bis positive, program flow continues to100e, at which the program identifies point C(p) to the user and then can require the user to point to C(p) usingpointing device20, advantageously (but not necessarily) aided by light-beam projection device202, preferably from a position such as point CA, as determined previously at90a(FIG. 4). The user can also be queried to affirm the pointing action by, for example, activating one of thebuttons203aor204b, or the like. The identification of point C(p) may occur, for example, by having the program display a visible screen mark521a-521dat a position on computerscreen interaction region51 associated with C(p), which may then be projected byprojection device40 to coincide with point C(p). Other means of identification are also possible, such as the characterization of point C(p) as ‘upper-right corner’ or the like.
• Program flow then continues on to100f, where the program can wait until the highlighting action is affirmed by the user, after which the 3D orientation of the z′-axis and the 3D position of the z′=0 point (of the z′-axis) are measured with respect to the x y z coordinate system, using coordinatesensing device201 and301, and communicated to the computer using communication andcontrol device204 and/or304. This 3D orientation and position is then associated with line A(p) and stored in memory. Program flow then continues to100c.
• After100cprogram flow continues to100dwhere a decision is made whether p is equal to n+1. If the decision is positive, it can be concluded that all necessary data on lines A(p) have been ascertained and program flow can continue to100g. If the decision at100dis negative, program flow reverts to100b.
• At100ga decision is made whether set B is empty. If the decision is positive, it can be concluded that enough information to uniquely determine the 3D coordinates of all the points in set P is contained in the a priori relationships combined with the available data on the lines in set A, and program flow continues to100p. If the decision at100gis negative, program flow continues to100h.
• At100hcounter p is again initialized to 1. The user is subsequently required to reposition thepointing device20 to a different location, displacing it substantially parallel toprojection region60 over a distance substantially equal to the distance betweenprojection region60 and his or her present location. Other locations may also be used.
• Flow then continues to100iwhere a decision is made whether line B(p) (connecting point CB to point C(p)) is included in set B. If the decision is negative, program flow continues to100jwhere counter p is incremented by 1.
• If the decision at100iis positive, program flow continues to100m, at which point the program identifies point C(p) to the user and can query the user to highlight this point using light-beam projection device202. The program can also query the user to affirm the highlighting by, for example, activating one of thebuttons203aor204bor the like. The identification of point C(p) may occur, for example, by having the program display avisible screen mark521a,521b, . . . at a position on computerscreen interaction region51 associated with C(p), which may then be projected byprojection device40 to coincide with point C(p). Other means of identification are also possible, such as the characterization of point C(p) as ‘upper-right corner’ or the like.
• Program flow then continues to100n, where the program may wait until the highlighting action is affirmed. After affirmation, the 3D orientation of the z′-axis and the 3D position of the point z′=0 (of the z′-axis) are measured with respect to the x y z coordinate system, using coordinatesensing device201 and301, and are communicated to the program using communication andcontrol device204 and/or304. The 3D orientation and position are then associated with line B(p) and can be stored in memory. Program flow then continues to100j.
• After100jprogram flow continues to100kwhere a decision is made whether p is equal to n+1. If the decision is positive, it is concluded that all necessary data for lines B(p) has been ascertained and program flow continues to100p. If the decision is negative, program flow reverts to100i.
• At100pit is concluded that the combined information of the a priori relationships and the data for lines in sets A and B that is stored in memory is enough to establish the coordinates of all points in P, as is further explained with reference toFIG. 6.
• Referring toFIG. 6, as will be appreciated by those skilled in the art, at110athe available information contained in the a priori relationships and the orientations and positions of the lines in sets A and B may be used to establish the coordinates of each of the points in set P, all measured with respect to the x y z coordinate system. It may happen that the 3D data associated with lines A(p) and B(p) (I<p<n+l) cause them not to intersect at point C(p); for example, this could be the case due to insufficient accuracy in determining the orientations and positions of lines A(p) and B(p). Under such circumstances point C(p) may be estimated or determined in such a way that it lies halfway along the shortest line-segment that connects lines A(p) and B(p). Other methods for determining point C(p) are also possible.
• Once the complete 3D description ofinteraction structure72 has been established (given the available data and assumptions), program flow continues to110b. At110b, a method M is constructed that maps points ofinteraction structure72 to computerscreen interaction region51. Such methods are well known to those skilled in the art. For example, a method such as that described in U.S. Pat. No. 6,373,961 to Richardson et al. (2002) may be used, but other appropriate methods may also be used. For completeness, the method disclosed in the Richardson et al. '961 patent will briefly be explained here. Referring toFIG. 18,interaction structure72 may be defined by line segments JK and ML. Furthermore, point F may define point-of-aim210, which is also the intersection between the z′-axis andinteraction structure72. The line segments JK and ML are generally not parallel. If not parallel, then the line segments JK, ML are extended until they meet at a point H. Then a line is drawn through F and H, that intersects segment JM at point P and segment KL atpoint 0. If the segments JK and ML are parallel, the a line is drawn instead through F that is parallel to JK, and which intersects the other two line segments as described. The process continues similarly with the other segments, which are extended to meet at point G. A line through F and G intersects segment JK at point R and segment ML at point Q. Subsequently, the following distances are measured and noted as percentages: JRJJK, KO/KL, LQ/LM and MP/MJ. Then four points S, T, U, V are chosen as shown inFIG. 18, that form an exact rectangle STUV; this rectangle may define computerscreen interaction region51. Further, points W, X, Y, Z are chosen on the sides of the rectangle STUV in such a way that SZ/ST=JRJJK, TW/TU=IKO/KL, UX/UV=LQ/LM and VYNS=MP/MJ. Then the points Z and X are joined by a line, and the points Y and W are joined by a line. The point of the intersection of these latter two lines is N, which is the point that results from the operation of method M.
• Referring once again toFIG. 6, after110b, program flow continues to110cwhere a decision is made whether the user has requested the program to end. If the decision is positive, the program ends at110k.
• If the decision at110cis negative, program flow continues to110d, where a decision is made regarding whether the user has requested a direct-pointing action, such as activatingbutton203aor203bor the like.
• If the decision at110dis negative, program flow continues to110e, where the light-beam projection device202 can be instructed or caused to de-activate (using, for example,control device204 and/or304). Furthermore, a software routine (not shown) that controls computer cursor501 (see alsoFIG. 2) is instructed that the user does not want to execute a direct-pointing action, after which program flow reverts to110c. Such cursor control routines are well known to those skilled in the art and are therefore not described here.
• If the decision at110dis positive, program flow continues to110f, where a decision is made whether the z′-axis intersectsinteraction structure72. Those skilled in the art will appreciate that this is possible because all relevant 3D information is known or is measurable by coordinatesensing device201 and301. If the decision is positive, program flow continues to110h, at which method M is used to map the intersection point of the z′-axis withinteraction structure72 to computerscreen interaction region51. This mapped position, as well as the user's desire to execute a direct-pointing action are communicated to the cursor control routine (not shown) running on the computer.
• After110h, program flow continues to110iwhere a decision is made whether the user wishes to execute an action associated with the position ofcursor501. For example, activatingbutton203aor203bor the like may indicate such a request. If the decision is negative, program flow reverts to110c.
• If the decision at110iis positive, program flow continues to110j, where the cursor control routine (not shown) is requested to execute the intended action. Such an action may, for instance, be a ‘double-click’ action or the like. Program flow then reverts to110c.
• There may be situations for which the system of equations that allows determination of all coordinates of the points in set P will be somewhat ill-posed. This could for example occur if locations CA and CB are chosen too close together, or if the angles of somepointing lines21 with respect tointeraction region71 are too small, as will be appreciated by those skilled in the art. In such cases, the user may be directed to choose a different point for locations CA and/or CB from where the various points C(p) are to be highlighted.
• Thus, methods and means are disclosed that afford a highly flexible and easily deployable system for interacting with a presentation in a direct-pointing manner at a location not specifically designed or adapted for such a purpose. Moreover, although it is desirable to have highly accurate coordinatesensing device201 and301, in some instances such may not be necessary because the visual feedback afforded by a displayed cursor will compensate to a large extent for any errors made in the measurements of coordinate parameters. The same holds true for the importance of any discrepancies between the actual shape ofinteraction region71 and that of theinteraction structure72 that is assumed to coincide with it. Also, minor discrepancies
• between the actual position and orientation ofinteraction region71 and the assumed position and orientation of interaction structure72 (which is assumed to coincide with interaction region71) need not be of critical importance, as will be appreciated by those skilled in the art.
• Mathematically, there are infinite possible shapes forinteraction structure72 as well as infinite a priori relationships, sets P, A, B and methods M. To further explain the invention, what follows is a list of examples that are believed will be encountered most often. These examples are not meant to restrict the scope of the present invention in any way, but merely serve as further clarification.
• Referring toFIG. 7, it can be assumed that the venue where the presentation is to be made is set up such thatprojection image70 andinteraction region71 are substantially rectangular. Such may be the case when the optical axis ofprojection device40 is substantially parallel to the normal toprojection region60.Projection region60 may be a projection screen on a stand, as shown, but may also be a wall or any other substantially vertical surface. It can also be assumed that the bottom and top edges of bothprojection image70 andinteraction region71 are substantially horizontal. Furthermore, it can be assumed thatinteraction region71 is substantially the same size asprojection image70 and that computerscreen interaction region51 is substantially the same size ascomputer screen image50. Additionally, it can be assumed that calibration points721a,721dsubstantially coincide with the projected versions of screen marks521a,521d. Finally, it is assumed thatbase station30 is positioned such that the x-y-plane is substantially horizontal and the x-z-plane substantially coincides with the plane ofinteraction region71. To facilitate in the positioning ofbase station30, the user may be aided by level-sensing device303 andvisible markings302.
• Referring toFIGS. 3 and 7, process elements80a-jmay then result, either through default settings or through user-supplied settings, in the assumptions thatinteraction structure72 is a rectangle that lies in the x-z plane and that its top and bottom edges are parallel to the x-y plane.
• Referring toFIGS. 4 and 7,program element90amay then result in a set P containing three points that define the upper-right corner C(l), the upper-left corner C(2) and the lower-left corner C(3) ofinteraction structure72. Additionally,program element90amay determine point CA (not denoted. as such inFIG. 7) to lie away fromprojection region60, for example at the center of one of the two instantiations of pointingdevice20, as shown inFIG. 7. Point CB may be determined anywhere in space.Program element90amay also result in sets A and B each containing three lines, connecting the points in set P to points CA and CB respectively.Program element90cmay then result in a priori relationships such that all points in set P have y-coordinate equal to zero, that the upper-left corner will have an x-coordinate equal to the x-coordinate of the lower-left corner and that its z-coordinate will be equal to the z-coordinate of the upper-right corner. It will then be appreciated by those skilled in the art that this a priori information, together with complete 3D information on two lines in set A will be sufficient to uniquely determine the position, size and orientation ofinteraction structure72 with respect to the x y z coordinate system. Therefore,program elements90d,90e,90f,90gmay result in an empty set B. Moreover,program elements90h,90i,90j,90kmay result in the removal from set A of line A(2), connecting point CA to point C(2).
• Referring toFIGS. 5 and 7, program elements100a-pmay result in the program identifying point C(1) by instructing the computer (not shown) andprojection device40 toproject screen mark521aontoprojection region60, resulting in the appearance ofcalibration point721a. Subsequently, the user is required to use light-beam projection device202 to highlightcalibration point721aand indicate the success of this action to the computer (not shown) by, for example, pressingbutton203a. As a result, the orientation and position of the z′-axis are assumed to define line A(1). The same actions are then performed for point C(3) and line A(3). Since set B is empty, the user need not be required to repositionpointing device20. The two depictions of pointingdevice20 inFIG. 7 are meant to illustrate that pointingdevice20 only needs to be redirected and not repositioned during this exercise. It will be appreciated by those skilled in the art thatprogram element100pwill then successfully result in a full 3D description of the three points in set P.
• Referring toFIGS. 6 and 7,program element110amay then result in the description ofinteraction structure72 as a rectangle with upper-right corner C(l), upper-left corner C(2) and lower-left corner C(3). Step110bmay then result in a method M that maps a point 6 (not shown) ininteraction structure72 to a point E (not shown) in computerscreen interaction region51 in such a way that the ratio of distances between 6 and any two of the four corners ofinteraction structure72 is the same as the ratio of distances between s and the two corresponding corners of computerscreen interaction region51, as will be appreciated by those skilled in the art.Steps110c-kmay then result in a very intuitive cursor control device that responds to a direct-pointing action by de-activating light-beam projection device202 and showingcursor501 at substantially point-of-aim210 of pointingdevice20 whenever point-of-aim210 lies in projection image70 (which in this example, by assumption, substantially coincides with interaction region71). Since there is no necessity for a computer generated cursor and a light spot to be visible simultaneously there will not be any cause for confusion as to which of the two is the ‘actual cursor’.Cursor501 may be hidden from view when point-of-aim210 does not lie ininteraction structure72, in which case light-beam projection device202 may be activated automatically. Also, intuitive actions such as ‘double click’ may be provided bysteps110c-k.
• Referring toFIG. 8, it can be assumed that the venue where the presentation is to be given is set up such thatprojection image70 andinteraction region71 are substantially rectangular. Such may be the case when the optical axis ofprojection device40 is substantially parallel to the normal toprojection region60.Projection region60 may be a projection screen on a stand, as shown, but may also be a wall or any similar surface.Projection region60 can be assumed to be oriented substantially vertically. It can also be assumed that the bottom and top edges of bothprojection image70 andinteraction region71 are substantially horizontal. Furthermore, it can also be assumed thatinteraction region71 is substantially the same size asprojection image70 and that computerscreen interaction region51 is substantially the same size ascomputer screen image50. Additionally, it can be assumed that calibration points721a,721c,721dsubstantially coincide with the projected versions of screen marks521a,521c,521d. Finally, it can be assumed thatbase station30 is positioned such that the xy-plane is substantially horizontal. No assumptions are made regarding the x-z-plane and the y-z-plane other than that they are substantially vertical. To facilitate in the positioning ofbase station30, the user may be aided by level-sensing device303.
• Referring toFIGS. 3 and 8, program elements80a-jmay then result, either through default settings or user-supplied settings, in the assumptions thatinteraction structure72 is a rectangle of which the top and bottom edges are parallel to the x-y-plane (i.e., they are horizontal) and the left and right edges are perpendicular to the x-y-plane (i.e., they are vertical).
• Referring toFIGS. 4 and 8,program element90amay then result in a set P containing three points that define the upper-left corner C(1), the upper-right corner C(2) and the lower-left corner C(3) ofinteraction structure72. Additionally,program element90amay determine point CA (not denoted as such inFIG. 8) to lie away fromprojection region60, for example, at the center of one of the three instantiations of pointingdevice20, as shown inFIG. 8. Point CB (not denoted as such inFIG. 8) may be determined to be displaced from point CA in a direction substantially parallel tointeraction region71, over a distance that may be the same order of magnitude as the size ofinteraction region71.Program element90amay also result in sets A and B each containing three lines, connecting the points in set P to points CA and CB respectively.Program element90cmay then result in a priori relationships requiring that the upper-left corner will have x- and y-coordinates equal to the x- and y-coordinates of the lower-left corner and that its z-coordinate will be equal to the z-coordinate of the upper-right corner. As will be explained in detail below, the a priori information together with complete 3D information on two lines in set A and one line in set B will be enough to uniquely determine the position, size and orientation ofinteraction structure72 with respect to the x y z coordinate system. Therefore,program elements90d,90e,90f,90gmay result in set B containing only line B(3). Moreover,program elements90h,90i,90j,90kmay result in set A containing only lines A(1) and A(2).
• Referring toFIGS. 5 and 8, program elements100a-pmay result in the program identifying point C(1), C(2) and C(3) by means ofprojection device40, in a manner similar to the one described in the previous example. In this case, however, C(1) and C(2) may be highlighted from approximately the same location CA, but C(3) needs to be highlighted from a different” location CB. To explain that the above-mentioned information is enough to uniquely establish the position, size and orientation ofinteraction structure72, define the points in set P as:
• $\begin{array}{cc}C\left(1\right)=\left(\begin{array}{c}{x}_1+{\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_1·{\mathrm{Rx}}_1\\ {\mathrm{<figure-callout\; label="y"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">y</figure-callout>}}_1+{\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_1·{\mathrm{Ry}}_1\\ {\mathrm{<figure-callout\; label="z"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">z</figure-callout>}}_1+{\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_1·{\mathrm{Rz}}_1\end{array}\right)& \left(1\right)\\ C\left(2\right)=\left(\begin{array}{c}{x}_1+{\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_2·{\mathrm{Rx}}_2\\ {\mathrm{<figure-callout\; label="y"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">y</figure-callout>}}_1+{\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_2·{\mathrm{Ry}}_2\\ {\mathrm{<figure-callout\; label="z"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">z</figure-callout>}}_1+{\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_2·{\mathrm{Rz}}_2\end{array}\right)& \left(2\right)\\ C\left(3\right)=\left(\begin{array}{c}{x}_3+{\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00015.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_3·{\mathrm{Rx}}_3\\ {\mathrm{<figure-callout\; label="y"\; filenames="US20190317613A1-20191017-D00015.png"\; state="\{\{state\}\}">y</figure-callout>}}_3+{\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00015.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_3·{\mathrm{Ry}}_3\\ {\mathrm{<figure-callout\; label="z"\; filenames="US20190317613A1-20191017-D00015.png"\; state="\{\{state\}\}">z</figure-callout>}}_3+{\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00015.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_3·{\mathrm{Rz}}_3\end{array}\right)& \left(3\right)\end{array}$
• Here, points CA and CB are defined as (x₁, y₁, z₁) and (x₃, y₃, z₃) respectively. Moreover, lines A(1), A(2) and B(3) are defined as passing through CA, CA and CB respectively, lying in directions governed by (Rx₁, Ry₁, Rz₁), (Rx₂, Ry₂, Rz₂) and (Rx₃, Ry₃, Rz₃). All of these quantities are presumed to be measured by coordinatesensing device201 and301. For a unique description of the points in set P a solution is required for λ₁, λ₂and λ₃.
• Using these definitions the conditions described above can be written as:
• $\begin{array}{cc}\left[\begin{array}{ccc}{\mathrm{Rz}}_1& -{\mathrm{<figure-callout\; label="Rz"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">Rz</figure-callout>}}_2& 0\\ {\mathrm{<figure-callout\; label="Rx"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">Rx</figure-callout>}}_1& 0& -{\mathrm{<figure-callout\; label="Rx"\; filenames="US20190317613A1-20191017-D00015.png"\; state="\{\{state\}\}">Rx</figure-callout>}}_3\\ {\mathrm{<figure-callout\; label="Ry"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">Ry</figure-callout>}}_1& 0& -{\mathrm{Ry}}_3\end{array}\right]\left(\begin{array}{c}{\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_1\\ {\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_2\\ {\lambda }_3\end{array}\right)=\left(\begin{array}{c}0\\ x_3-x_1\\ y_3-{\mathrm{<figure-callout\; label="y"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">y</figure-callout>}}_{1}\end{array}\right)& \left(4\right)\end{array}$
• which can be solved in a straightforward manner according to the expression:
• $\begin{array}{cc}\left(\begin{array}{c}{\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_1\\ {\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_2\\ {\lambda }_3\end{array}\right)=\left(\begin{array}{c}\frac{\left[{\mathrm{<figure-callout\; label="Rx"\; filenames="US20190317613A1-20191017-D00015.png"\; state="\{\{state\}\}">Rx</figure-callout>}}_3·\left(y_3-y_1\right)+{\mathrm{<figure-callout\; label="Ry"\; filenames="US20190317613A1-20191017-D00015.png"\; state="\{\{state\}\}">Ry</figure-callout>}}_3·\left(x_1-x_3\right)\right]}{\left[{\mathrm{<figure-callout\; label="Rx"\; filenames="US20190317613A1-20191017-D00015.png"\; state="\{\{state\}\}">Rx</figure-callout>}}_3·{\mathrm{Ry}}_1-{\mathrm{<figure-callout\; label="Rx"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">Rx</figure-callout>}}_1·{\mathrm{Ry}}_3\right]}\\ \frac{\left[{\mathrm{<figure-callout\; label="Rz"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">Rz</figure-callout>}}_1/{\mathrm{Rz}}_2\right]·\left[{\mathrm{<figure-callout\; label="Rx"\; filenames="US20190317613A1-20191017-D00015.png"\; state="\{\{state\}\}">Rx</figure-callout>}}_3·\left(y_3-y_1\right)+{\mathrm{<figure-callout\; label="Ry"\; filenames="US20190317613A1-20191017-D00015.png"\; state="\{\{state\}\}">Ry</figure-callout>}}_3·\left(x_1-x_3\right)\right]}{\left[{\mathrm{<figure-callout\; label="Rx"\; filenames="US20190317613A1-20191017-D00015.png"\; state="\{\{state\}\}">Rx</figure-callout>}}_3·{\mathrm{Ry}}_1-{\mathrm{<figure-callout\; label="Rx"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">Rx</figure-callout>}}_1·{\mathrm{Ry}}_3\right]}\\ \frac{\left[{\mathrm{<figure-callout\; label="Rx"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">Rx</figure-callout>}}_1·\left(y_3-y_1\right)+{\mathrm{<figure-callout\; label="Ry"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">Ry</figure-callout>}}_1·\left(x_1-x_3\right)\right]}{\left[{\mathrm{<figure-callout\; label="Rx"\; filenames="US20190317613A1-20191017-D00015.png"\; state="\{\{state\}\}">Rx</figure-callout>}}_3·{\mathrm{Ry}}_1-{\mathrm{<figure-callout\; label="Rx"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">Rx</figure-callout>}}_1·{\mathrm{Ry}}_3\right]}\end{array}\right)& \left(5\right)\end{array}$
• Note that this solution shows that, if point C(3) was highlighted from any point on thepointing line21 that is used to highlight point C(1), i.e.,
• $\begin{array}{cc}\left(\begin{array}{c}{x}_3\\ y_3\\ z_3\end{array}\right)=\left(\begin{array}{c}{x}_1+\tau ·{\mathrm{Rx}}_1\\ {\mathrm{<figure-callout\; label="y"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">y</figure-callout>}}_1+\tau ·{\mathrm{Ry}}_1\\ {\mathrm{<figure-callout\; label="z"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">z</figure-callout>}}_1+\tau ·{\mathrm{Rz}}_1\end{array}\right)& \left(6\right)\end{array}$
• the solution for the three unknown A's would collapse to
• $\begin{array}{cc}\left(\begin{array}{c}{\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_1\\ {\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_2\\ {\lambda }_3\end{array}\right)=\left(\begin{array}{c}\tau \\ \tau {\mathrm{Rz}}_1/{\mathrm{<figure-callout\; label="Rz"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">Rz</figure-callout>}}_2\\ 0\end{array}\right)& \left(7\right)\end{array}$
• making unique determination ofinteraction structure72 impossible. Conversely, the fact that points C(1) and C(2) are, in this example, both highlighted from substantially the same point CA does not cause any problems. It will be appreciated by those skilled in the art that C(1) and C(2) may also be highlighted from different points, without loss of functionality.
• Referring toFIGS. 6 and 8, the course and results of program elements110a-kwill be similar to those described in Example 1.
• Referring toFIG. 9, in another example, the only assumptions made are thatinteraction region71 is substantially the same size asprojection image70, computerscreen interaction region51 is substantially the same size ascomputer screen image50, calibration points721a,721b,721c,721dsubstantially coincide with the. Projected versions of screen marks521a,521b,521cand521dand computerscreen interaction region51 is substantially rectangular. This situation may arise when the optical axis ofprojection device40 is not aligned with the normal toprojection surface60.
• Referring toFIGS. 3 and 9, program elements80a-jmay then result, either through default settings or user supplied settings, in the assumption thatinteraction structure72 is a quadrangle; no assumptions are made regarding its position or orientation.
• Referring toFIGS. 4 and 9,program elements90amay then result in a set P containing four points that define the upper-right, upper-left, lower-right and lower-left corners ofinteraction structure72. Additionally,program element90amay conceive point CA (not denoted as such inFIG. 9) to lie away fromprojection region60, for example at the center of the first of the two instantiations of pointingdevice20, as shown inFIG. 9. Point CB may, for example, be determined to lie at the center of the second of the two instantiations of pointingdevice20, as drawn inFIG. 9. Now,program elements90d,90e,90f,90gand steps90h,90i,90j,90kmay result in sets A and B each containing 4 lines.
• Referring toFIGS. 5 and 9, program elements100a-pmay result in the program identifying the four points in set P by means ofprojection device40, in a manner similar to the one described in Example 1. In this case, however, all four points in set P may be highlighted from approximately the same location CA first, after which all four points may be highlighted from approximately the same location CB. Note thatFIG. 9, for reasons of clarity, only depicts the highlighting ofcalibration point721a. Each point C(i) may then be constructed as the point lying halfway the shortest line segment connecting lines A(i) and B(i); this line segment may have zero length.
• Referring toFIGS. 6 and 9, the course and results of program elements110a-kwill be similar to those described with respect to the first example, with the exception ofprogram element110b, the mapping element. Now, a more elaborate method M is needed than the one described in previous examples. For example, a method such as described in U.S. Pat. No. 6,373,961 to Richardson (2002) may be utilized, but other appropriate methods may also be used.
• A second embodiment will now be made with reference toFIG. 10, which showspointing device20 andbase station30. Here, in addition to any or all of the elements mentioned in the previous embodiment, pointingdevice20 is also provided withdistance measuring device206, the position and orientation of which with respect to the x′ y′ z′ coordinate system may be ascertained at the place of manufacture thereof and will be presumed to be known. Distance measuringdevice206 may for example be embodied by a focus adjustment, or optical sensor. A digital camera may also be used. Distance measuringdevice206 may determine the point-of-aim-distance211 between the origin of the x′ y′ z′ coordinate system and the point-of-aim, measured substantially parallel to the z′-axis (see alsoFIG. 2). The point ofaim210 may lie onprojection region60, on whichinteraction region71 may lie. Distance measuringdevice206 may have associated with it a manual focus adjustment that may be adjusted by the user (manual focus adjustment not shown). Alternatively,distance measuring device206 may comprise a circuit (not shown) that automatically determines the point-of-aim-distance211 between the origin of the x′ y′ z′ coordinate system and the point-of-aim. For example, if the point-of-aim lies onprojection region60,distance measuring device206 may bounce light off ofprojection region60. In doing so, use may for instance be made of light-beam projection device202. Any other appropriate mechanism for determining distance may also be used fordistance measuring device206. The optical axes of light-beam projection device202 anddistance measuring device206 may coincide, for instance by making use of partially-reflecting elements (not shown) such as disclosed in U.S. Pat. No. 4,768,028 to Blackie (1988).
• The operation of the present embodiment will now be described with reference toFIGS. 2, 10, and 3. The process elements inFIG. 3 will be identical to the ones described in the first embodiment, resulting in similar assumptions regarding the shape of interaction region71 (and interaction structure72) and possibly orientation and position ofinteraction structure72. In particular, a default distance betweeninteraction structure72 and the origin of coordinate system x y z will be assumed for purpose of initialization.
• Referring toFIG. 11, instead of the program elements described above with reference toFIG. 4, program flow now continues at120a. Program elements120a-120hare shown to be similar to program elements90a-90m(FIG. 4), except that only point CA, set A and counter a are considered. That is to say, in the present embodiment it is no longer necessary to determine the second point CB and associated repositioning ofpointing device20 during the calibration procedure.
• Afterprogram element120h, program flow continues with program elements outlined inFIG. 12. Referring toFIG. 12, program elements130a-130gare seen to be similar to steps100a-100p(FIG. 5). Comparingprogram element130fwithprogram element100f(FIG. 5), it can be seen that at130fthe program also stores information on the point-of-aim-distance (211 inFIG. 2) between the origin of the x′ y′ z′ coordinate system and at least one of points C(p). Comparingelements130gand100p(FIG. 5), it can be seen that these point-of-aim-distances (211 inFIG. 2) are also used in determining the 3D coordinates of the points in set P.
• With complete information on the 3D coordinates of the points in set P, it is then possible to construct a 3D description of interaction structure72 (inFIG. 2). Therefore, program elements110a-110kas described above with reference toFIG. 6 may also be followed in the present embodiment.
• Thus, methods and means are disclosed that afford a highly flexible and easily deployable system for interacting with a presentation in a direct-pointing manner at a location not specifically equipped for such a purpose. Moreover, when utilizing the second preferred embodiment it is often sufficient for the user to highlightvarious calibration points721a,721b, . . . from one and the same position, making the calibration procedure even more easy to follow.
• Another embodiment will now be explained with reference toFIG. 13.FIG. 13 shows another embodiment of thepointing device20. Pointingdevice20 may be equipped with any or all of the elements mentioned in the other embodiments, such as described above with reference toFIG. 1 andFIG. 10. The present embodiment may include abase station30 that has associated with it a coordinate system x y z. However, in the present embodiment it can be assumed that the position of pointingdevice20 will remain substantially unchanged relative tointeraction region71 over an extended period of time, and that the origin of coordinate system x y z may be assumed to be virtually anywhere instead of being related to thebase station30. Furthermore, if any of the axes of the x y z system are chosen to be related to (locally) well-established, independent and stationary directions such as, for example, the Earth's magnetic field and/or the Earth's gravitational field, the Earth itself may be interpreted as embodyingbase station30. Under such circumstances there may not be a need for anartificial base station30; coordinatesensing device201 may then, for example, be embodied by devices sensing the directions of the Earth magnetic and gravitational fields, such as accelerometers and a compass (for example, model no. HMR3300 device as manufactured by Honeywell International Inc., Morristown, N.J.), and communication andcontrol device204 may be configured to communicate directly with the computer (not shown).
• The operation of the present embodiment will now be described with reference toFIGS. 8, 13, and 3. The remainder of the description of the present embodiment will, for purposes of explanation, include the assumption that any separately embodied base station30 (FIGS. 8 and 13) is omitted entirely. It is also assumed that one of the axes of the orthogonal coordinate system x y z is substantially parallel to the Earth's gravity field (or has a defined relationship thereto), while the second and third axes of the coordinate system are in a fixed, known relationship to the at least one geographically defined axis. In the present embodiment it is assumed that the relative position of pointingdevice20 with respect tointeraction region71 remains substantially unchanged, therefore the origin of coordinate system x y z will, for purposes of explanation, be chosen to coincide with a fixed point in thepointing device20. For the purpose of the present embodiment, it should be understood that the three instantiations of pointingdevice20 as shown inFIG. 8 substantially coincide.
• It can be assumed in the present embodiment that a display system is arranged at the venue where a presentation is to be made, as a combination of aportable projection device40 andprojection surface60, for example a wall. It will furthermore be assumed that this display system is connected, using appropriate means, to the computer (not shown) that generates the presentation imagery.
• Upon arriving at the venue, the user connectspointing device20 to the computer (not shown), for example via a USB connection (not shown), or using a wireless relay device (not shown). Also, the computer may recognize the connection and start a program, part of which may be contained in communication andcontrol device204 or in control logic contained in the wireless relay device itself (not shown). Alternatively, the user may be required to load the program into the computer manually via a CD drive, a floppy drive, memory stick or the like (not shown). In any case, the program may initiate a calibration routine that has as its object establishing the shape, position, size and orientation of a well-definedinteraction structure72, relative to the x y z coordinate system, wherein theinteraction structure72 is assumed to substantially coincide with a scaled and parallel version ofinteraction region71. The flow of this program will be explained with reference toFIG. 3.
• At80athe program is initiated. At80bdefaults are entered forinteraction region71. Specifically,interaction region71 is assumed to be a substantially parallel and scaled version of a well-definedinteraction structure72. Moreover, the respective corners ofinteraction region71 andinteraction structure72 are assumed to substantially lie on lines intersecting each other in the origin of coordinate system x y z. It should be noted thatFIG. 8 shows aninteraction region71 and aninteraction structure72 that are almost coincident, but this is only meant for illustrative purposes and is not a limitation on the scope of the invention. Again referring toFIG. 3, at80bdefault values are also established for the orientation and position of theinteraction structure72. For example, the default values may provide thatinteraction region71 is substantially a parallel and scaled version of aninteraction structure72 that is a flat rectangle of which the two most vertical sides are substantially parallel to the Earth's gravitational field, and of which the two most horizontal sides are substantially perpendicular to the Earth's gravitational field; moreover, the default values may provide that the respective corners ofinteraction region71 andinteraction structure72 substantially lie on lines intersecting each other in the origin of coordinate system x y z. The default values may furthermore provide that the position ofinteraction structure72 is such that the distance betweenpointing device20 and the lower left corner ofinteraction structure72 is equal to 1. Note that other values for the foregoing distance may be equally valid in the present embodiment. Based on the foregoing default values, calibration points721a,721c,721d, . . . can define characteristic features ofinteraction region71 and can also define characteristic features ofinteraction structure72.
• At80ca decision is made whether the default values forinteraction region71 andinteraction structure72 should be accepted or overridden by the user. If the default values are to be accepted, the program flow continues to80j, the details of which are explained below with reference toFIG. 11. If the defaults are to be overridden, the program flow continues withelements80d-80i, during which the user may override any of the default settings. The user may be aided duringprogram elements80d-80iby a library of predetermined shapes, predetermined orientations and/or predetermined positions. Alternatively, the user can be provided with the capability to construct custom shapes, orientations and/or positions. In any event, program flow continues to80j.
• Referring toFIG. 11, program flow now continues to120a. Program elements120a-120hare substantially similar to program elements90a-90m(FIG. 4), except that only point CA, set A and counter a. are considered. In the present embodiment it is not required to determine a second point CB and an associated repositioning ofpointing device20 during the calibration procedure.
• After120h, program flow continues with elements described with reference toFIG. 14. InFIG. 14, program elements140a-140gare shown to be similar to steps130a-130g(FIG. 12). Comparingelement140fwithelement130f(FIG. 12), it can be seen thatelement140fdoes not include storing information for any directly measured distance.Element140fneed not include storing information on the position of the z′-axis, because the position of pointingdevice20 is assumed to remain substantially unchanged over an extended period of time. Comparingelements140gand130g, it can be seen that any directly measured distance is not needed to determine the 3D coordinates of the points in set P (seeFIG. 4). In fact, having the default distance (set to 1) betweenpointing device20 and the lower left corner ofinteraction structure72 is sufficient for operation of the system in the present embodiment. To explain this fact, reference is again made toFIG. 8. Using the above default values,interaction structure72 will be completely defined by, for example, its upper-left corner C(1), its upper-right corner C(2) and its lower-left corner C(3)
• $\begin{array}{cc}C\left(1\right)={\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_1·\left(\begin{array}{c}\mathrm{Rx}1\\ \mathrm{Ry}1\\ \mathrm{Rz}1\end{array}\right)& \left(8\right)\\ C\left(2\right)={\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_2·\left(\begin{array}{c}\mathrm{Rx}2\\ \mathrm{Ry}2\\ \mathrm{Rz}2\end{array}\right)& \left(9\right)\\ C\left(3\right)={\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00015.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_3·\left(\begin{array}{c}\mathrm{Rx}3\\ \mathrm{Ry}3\\ \mathrm{Rz}3\end{array}\right)& \left(10\right)\end{array}$
• lying on lines A(1), A(2) and A(3) that connect the origin to points (Rx₁, Ry₁, Rz₁), (Rx₂, Ry₂, Rz₂) and (Rx₃, Ry₃, Rz₃) respectively. Using these definitions, the conditions described above determine relationships between the various variables that can be written as
• $\begin{array}{cc}\left[\begin{array}{ccc}{\mathrm{Rz}}_1& -{\mathrm{<figure-callout\; label="Rz"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">Rz</figure-callout>}}_2& 0\\ {\mathrm{<figure-callout\; label="Rx"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">Rx</figure-callout>}}_1& 0& -{\mathrm{<figure-callout\; label="Rx"\; filenames="US20190317613A1-20191017-D00015.png"\; state="\{\{state\}\}">Rx</figure-callout>}}_3\\ {\mathrm{<figure-callout\; label="Ry"\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">Ry</figure-callout>}}_1& 0& -{\mathrm{Ry}}_3\end{array}\right]\left(\begin{array}{c}{\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_1\\ {\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_2\\ {\lambda }_3\end{array}\right)=\left(\begin{array}{c}0\\ 0\\ 0\end{array}\right)& \left(12\right)\end{array}$
• which will have non-trivial solutions only if the determinant of the matrix equals zero. Then, it can be shown that any combination of (λ₁, λ₂, λ₃) that can be written in the form:
• $\begin{array}{cc}\left(\begin{array}{c}{\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_1\\ {\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_2\\ {\lambda }_3\end{array}\right)=\left(\begin{array}{c}\mathrm{Rz}2/\mathrm{Rz}1\\ 1\\ \left(\mathrm{Rz}2·\mathrm{Rx}1\right)/\left(Rz1·\mathrm{Rx}3\right)\end{array}\right)·\alpha & \left(13\right)\end{array}$
• where α is any arbitrary real number, will generate solutions forinteraction structure72 that are parallel to each other. The assumption that the distance betweenpointing device20 and the lower left corner ofinteraction structure72 is 1 will hence uniquely generate one of these solutions. To see that other assumptions regarding this distance do not influence the operation of the present embodiment, reference is made toFIG. 15. InFIG. 15, projections of two parallel solutions forinteraction structure72 are shown, obtained for different values of distances CA-C(3). When it is assumed, for purposes of explanation, that these two solutions are parallel to the z-x plane,FIG. 15 may be interpreted as a projection of the two solutions onto the x-y plane. These projections are shown as the thick line segments C(3)-C(2) and CC3-CC2.FIG. 15 also shows the projection of point CA (i.e., the origin of coordinate system x y z), from where the lower left and upper right corners ofinteraction structure72 were highlighted, and line-segments C(3)-CA and C(2)-CA which coincide with the (projections of) the highlighting lines A(3) and A(2) (not denoted as such inFIG. 15). Finally, line-segment BB1-CA indicates (the projection of) a pointing line that is consistent with a point-of-aim BB1 on the one and point-of-aim BB2 on the other of the two solutions forinteraction structure72. To show that BB1 and BB2 represent the same horizontal coordinate when measured relative to the appropriate solution forinteraction structure72, the lengths of line-segments C(3)-BB1, C(2)-BB1, CC3-BB2, CC2-BB2, BBl-CA and BB2-CA are represented by bb1, bb2, b1, b2, aa12 and a12, respectively. It is sufficient to show that bb1/bb2=b1/b2. Consider the following equalities:
• $\begin{array}{cc}\frac{\mathrm{bb}1}{\sin \mathrm{\beta 1}}=\frac{\mathrm{aa}12}{\sin \alpha 1}& \left(14\right)\\ \frac{\mathrm{bb}2}{\sin \beta 2}=\frac{\mathrm{aa}12}{\sin \alpha 2}& \left(15\right)\end{array}$
• while, at the same time
• $\begin{array}{cc}\frac{b1}{\sin \beta 1}=\frac{a12}{\sin \mathrm{\alpha 1}}& \left(16\right)\\ \frac{b2}{\sin \beta 2}=\frac{a12}{\sin \alpha 2}& \left(17\right)\end{array}$
• From this it follows that
• $\begin{array}{cc}\frac{\mathrm{aa}12}{\mathrm{bb}1}=\frac{a12}{b1}& \left(18\right)\\ \frac{\mathrm{aa}12}{\mathrm{bb}2}=\frac{a12}{b2}& \left(19\right)\end{array}$
• which, in turn, provides that
• $\begin{array}{cc}\mathrm{bb}1\frac{a12}{b1}=\mathrm{bb}2\frac{a12}{b2}\Rightarrow \frac{\mathrm{bb}1}{\mathrm{bb}2}=\frac{b1}{b2}& \left(20\right)\end{array}$
• This implies that BB1 and BB2 represent the same horizontal coordinate when measured relative to the appropriate solution forinteraction structure72. Note that, ifFIG. 15 is interpreted as a projection onto the z-y-plane, it can be concluded that BB1 and BB2 also represent the same vertical coordinate when measured relative to the appropriate solution ofinteraction structure72. Therefore the initial assumption that the distance between C(3) and CA is equal to 1 does not influence the operation of the third preferred embodiment.
• It is theoretically possible that there are no non-zero solutions for the parameters λ₁, λ₂and λ₃, when the determinant referred to above in equation (12) is non-zero. Such may be the case, for example, due to errors in measurements or in the assumed rectangular shape ofinteraction region71. In such cases additional techniques may be used to find an acceptable solution forinteraction structure72. For example, a minimization routine may be used to find a solution forinteraction structure72 that minimizes a summation of the distances between some of its corners (which, by assumption, lie on lines connecting the origin of coordinate system x y z with the corners of interaction region71) and the highlighting lines.
• With complete information on the 3D coordinates of C(1), C(2) and C(3) it is possible to construct a 3D description ofinteraction structure72. Therefore, program elements110a-110kas described in the first embodiment and explained above with reference toFIG. 6 may also be followed in the present embodiment.
• There may be situations in which the user is not able to holdpointing device20 continuously at precisely the same position while performing the actions required by the calibration procedure described inFIGS. 3, 11 and 14, or during the use of the system as a direct-pointing device. Such changes in position of thepointing device20 can cause errors in the calculation of the point-of-aim. It will be appreciated that if the discrepancy between the calculated and the true point-of-aim is small enough, the visual feedback provided by the projection of the cursor (501 inFIG. 2; projected cursor not explicitly shown) during use of the system as a direct-pointing device will still afford the user the perception that direct-pointing is being performed.
• There are many other methods capable of establishing position, size and orientation ofinteraction structure72, as will be appreciated by those skilled in the art. For example, if the distance betweenpointing device20 andinteraction structure72 is presumed known, ifinteraction structure72 is assumed to be rectangular with two vertical and two horizontal sides and if its aspect ratio (the ratio between its horizontal size and its vertical size) is also presumed known, then knowledge of two lines from CA to the upper-right and lower-left corner is sufficient to narrow the number of solutions forinteraction structure72 down to 2, dictated by 2 solutions of a quadratic equation (see also paragraph181). One further assumption is then required to determine which of these 2 solutions is the correct one; this assumption may be in the form of a third line passing through yet another characteristic point ofinteraction structure72, but it may also be in the form of knowledge of the (approximate) angle between the plane in whichinteraction structure72 lies and a line connecting the origin to one of its corners. Other scenarios may also be conceived for which solutions may be devised that are within the scope of the general methods set forth herein.
• In the present embodiment, using well-established and stationary directions such as, for example, the Earth's magnetic field and the Earth's gravitational field, it may be possible to omit a separately embodied base station, making the present embodiment of the system possibly more compact, less expensive to make and easier to deploy.
• While the above description contains many specificities, these should not be construed as limitations on the scope of the invention, Many other variations are possible. For example, pointingdevice20 may be hand held, but may also be carried by the user in a different manner, such as by means of a headset, finger-worn ring or the like.
• Although the first, second and third embodiments make use of the assumption that theinteraction structure72 andinteraction region71 are quadrangles, this need not be the case even when the corresponding computerscreen interaction region51 is rectangular in shape. For example,projection region60 may be spherical in shape. This may cause the projection of a square computerscreen interaction region51 to result ininteraction region71 having the general shape of a sphere-segment. Other shapes are also possible, depending on a number of factors such as the angle between the optical axis ofprojection device40 andprojection region60. It will be appreciated by those skilled in the art that set P (seeFIG. 4, step90aandFIG. 11, step200a) and method M (seeFIG. 6, step110b) may become more complicated under these circumstances, but may still be well defined. In fact, method M and sets P, A, B may be constructed in any way that is advantageous to the particular system application. It is also possible to determine more points and associated line sets than just CA and CB, and associated line sets A and B. Specifically, set P may have more or fewer than the number of points described with reference to the foregoing embodiments, depending on the complexity of the presentation venue. Also, sets A and B may contain more lines than the minimum number needed to establish the coordinates of the points in set P.
• The present invention may make use of mathematical techniques not described explicitly herein but well known to those skilled in the art to aid with various aspects of the present invention, for example in the determination of position and/or orientation ofinteraction structure72. The second example of the first embodiment, for example, was described as relying on three lines highlighting corners of theinteraction structure72. It is also possible that a more accurate solution for the position and/or orientation ofinteraction structure72 may be obtained when the user is requested to also highlight the fourth corner of the screen. An appropriate minimization routine may then be used to find a solution characterized by, for example, a minimum accumulated distance between the four corners ofinteraction structure72 and the closest highlighting line.
• Moreover, although set P is described as containing points that are needed to completely defineinteraction structure72, other embodiments of the invention may use so-called “control points” for quality control purposes. Such control points may be used to test the validity of the assumptions made concerning, for example, the shape, the position, size and/or orientation ofinteraction region71 and, therefore,interaction structure72, as well as to test the accuracy of the measurements made to establish the 3D features of the various lines. As an example, consider the case in whichinteraction structure72 is taken to be a rectangle for which all data needed to uniquely establish its size, orientation and position with respect to the x y z coordinate system has been ascertained by the program elements described with respect to the foregoing embodiments. Moreover, in thisexample interaction region71 and, therefore,interaction structure72, are assumed to substantially coincide with the projection of a rectangular computerscreen interaction region51. Then, the computer may display asuitable screen mark521a,521b, . . . at, for example, the center of computerscreen interaction region51, which may be projected byprojection device40 ontoprojection region60. It will be appreciated that the process elements described herein may be used to calculate 3D coordinates of this projected point, based on the available data regarding shape, size, orientation and position ofinteraction structure72. The user may then be required to highlight the projected point using light-beam projection device202 and indicate the success of this action to the computer by, for example, activatingbutton203a. If all measurements and highlighting actions were accurate enough and no false assumptions were made, the 3D coordinates of the projected point should substantially conform to the 3D characteristics of the z′-axis. If the discrepancy is deemed too large the user may be required to repeat part or all of the calibration steps outlined in the present invention.
• There may also be more than one method M and sets P, A and B, each of which may be associated with adifferent interaction structure72 andinteraction region71. There may be more than oneinteraction structure72 andinteraction region71, each of which may, but need not, lie onprojection region60 and each of which may, but need not, be associated by means of projecting screen marks521a,521b, . . . , with one or more computerscreen interaction regions51.
• Although coordinatesensing devices201 and301 (see for exampleFIG. 1) were in some embodiments described as being able to provide information on the 3D position and 3D orientation of pointing device20 (see for exampleFIG. 2) relative to the x y z coordinate system, it should be understood that such information provision requirements may be relaxed under some circumstances. Specifically, in the first embodiment it is only required that the 3D position and 3D orientation of pointing line21 (see for exampleFIG. 2) be known, instead of having complete information on the position of pointingdevice20 along the pointingline21. That is, in somecases pointing device20 may be moved along pointingline21 without loss of functionality. In such embodiments the coordinate sensing devices need only provide information on the 3D position and 3D orientation of a line (as opposed to a line-segment) substantially intersectingpointing device20. If coordinate sensing devices are able to provide complete 3D information (i.e., 3D position and 3D orientation) ofpointing device20, as opposed to pointingline21, the extra positional information may, for example, be used to implement the capability to zoom in on a region around point-of-aim210 (not explicitly shown inFIG. 2); other actions may also be based on the extra positional information. The foregoing capability may be inferred from the description of the third embodiment, in which the coordinate sensing device is only used to provide information on the angle between the pointing line and two other fixed lines, instead of full position and orientation information about a line segment.
• Although the first, second and third embodiments describe the application of the invention as a cursor control device, the invention may also be used to facilitate non-cursor-related applications. Such applications may include “virtual writing” on theinteraction region71, target practice and the like. Also, the invention may be used other than for presentations such as, for example, to control a cursor on a television screen in order to make selections from menus associated with such things as Digital Satellite Television; other applications may also be anticipated.
• Moreover, the features of the present invention that enable the tracking of a point-of-aim relative to art interaction region may be enhanced by algorithms that allow the filtering of involuntary, fast and/or small hand-movements (that may be caused, for example, by the activation of buttons). Such algorithms may be used to control a point that moves less erratically than the actual point-of-aim while still being associated with it. Such motion filtering may produce a more steady motion ofcursor501. Such algorithms may also be used when the user is required to highlight calibration points721a,721b, . . . . Filter algorithms are well-known in the art.
• The present invention also contemplates the inclusion into pointingdevice20 of sensors capable of detecting motion or acceleration, both linear and angular, in addition to or as part of coordinatesensing device201. Such sensors may be used to detect unintended motion or acceleration of pointingdevice20 that may be caused, for example, by trembling of the user's hand. The programs described above with reference toFIGS. 5, 6, 12 and 14 may be enhanced by algorithms that compare the output of such sensors and/or coordinate sensing device to predefined thresholds, so that a change in measured orientation and/or position by coordinatesensing device201 may be recognized as significant only if such output exceeds these thresholds. Using threshold selection, a change in cursor position may only be affected if inferred to be intentional, avoiding the type of “jitter” associated with the use of regular laser pointers by a user with an unsteady hand.
• Furthermore, pointingdevice20 may be equipped with an image capturing device, such as a digital camera, preferably with zoom and controllable focus capabilities. Other image capturing devices, in combination with appropriate optical devices, may also be used. As stated before, such a digital camera may be used as an embodiment ofdistance measuring device206. Referring toFIG. 16, such a digital camera or the like (not shown) may also be used to aid in the process of highlighting the calibration points721a,721b, . . . . Images of calibration points721a,721b, . . . and light spots at point-of-aim210 may be captured during the time when the user is engaged in the highlighting process. These images may be temporarily stored, together with the position and orientation of the z′-axis at the time the image was acquired. During the highlighting process, it is likely that point-of-aim210 changes slightly. The resulting sequence of images may then be used to more accurately estimate the orientation and position of thepointing line21 that connects the origin of the x′ y′ z′ system tocalibration point721a,721b, . . . . For example, an averaging algorithm may be used. Alternatively, the stored position and orientation of the z′-axis corresponding to the captured image for which the light spot at point-of-aim210 is closest to the center ofcalibration point721a,721b, . . . may be taken as the position and orientation of thepointing line21 that connects the origin of the x′ y′ z′ system tocalibration point721a,721b, . . . .
• Some features may also be omitted without affecting the overall applicability of the present invention. Referring toFIG. 1, for example, level-sensing device303 andvisible markings302 may be helpful but are not of critical importance and may be omitted. If level-sensing device303 is present, the output may also be utilized to correct the orientation of the x y z coordinate system to ensure that its x-y plane is substantially horizontal, as previously explained. In such a case, thevisible markings302 may no longer precisely indicate the position of certain coordinate planes or even the origin of the x y z coordinate system. If the orientation of the uncorrected x-y-plane with respect to a horizontal surface is small enough, though,visible markings302 may still provide sufficiently reliable information on coordinate planes and origin of the corrected x y z coordinate system.
• Furthermore, the programs outlined inFIGS. 3, 4, 5, 6, 11, 12 and 14 may be changed appropriately without loss of functionality, particularly with respect to the order of some of the elements. The programs may also be written in any language that is advantageous to particular the implementation, including without limitation C, Java, and VisualBasic.
• Other embodiments that include automatic activation and de-activation of light-beam projection device202, or include provision for instructing the cursor control routines to show or hidecomputer cursor501 are also within the scope of the present invention. For example, automatic activation and de-activation of light-beam projection device202 may occur depending on whether or not the z′-axis intersects one of theinteraction structures72, or regions of space close to them. This may be performed by instructions to the cursor control routines to show or hidecomputer cursor501. Also, means may be provided to activate light-beam projection device202 manually.
• Furthermore, pointingdevice20 may include a conventional, indirect pointing device such as a trackball or the like, to be used in situations where direct pointing is not possible or not desired. As a practical matter, coordinatesensing device201, in some embodiments in combination with coordinatesensing device301, may be used to effect cursor control in an indirect manner if the intended use makes this desirable. To effect cursor control in indirect pointing applications, pointingdevice20 may include a user input device that enables the user to select whether indirect cursor motion control is desired, or may include sensors and/or algorithms that enable determining when direct pointing actions are not possible. For example, an algorithm or sensor may be included that enables determination of whether pointingdevice20 is within the operational range ofbase station30. The third described embodiment, specifically, may be enhanced with algorithms and/or sensors, such as accelerometers, that enable to determination of whether pointingdevice20 has been displaced significantly from the position at which the calibration procedure was performed, making direct pointing actions no longer possible without additional calibration steps. As part of the third described embodiment, additional input devices and algorithms may be included that enable the user to mark an orientation in space such that indirect cursor motion control may be effected based on the changes in orientation with respect to this marked orientation, as described in the cited prior art. Other embodiments of a system may also be enhanced by such functionality and input devices.
• In addition to the previously described methods to determine size, position and orientation ofinteraction region71, other methods are also within the scope of the invention. These include methods based on the use of digital cameras and the like. The position and orientation of such cameras, relative to the x y z coordinate system, may be tracked by using a device such as coordinatesensing device201 and301. For example, given sufficient knowledge of the optical characteristics of a digital camera, 3D features of an object may be determined from one or more images taken from one or more positions. Using such images, a complete 3D description ofinteraction region71 may be determined.
• Another alternative for establishing JD position, size and orientation ofinteraction region71 is provided by a digital camera or the like used in addition to, or as embodiment of, thedistance measuring device206. In such embodiments the direction relative to the x′ y′ z′ coordinate system of the axis along which distance is measured may be controlled and measured by mounting the digital camera and/ordistance measuring device206 in such a way that their orientation relative to the x′ y′ z′ coordinate system may be controlled and measured. Other implementations in which the foregoing components are positionally fixed are also contemplated. For purposes of explanation, the axis along which distance is measured is denoted as the z″-axis, and the z″=0 position is assumed to coincide with the origin of the x′ y′ z′ coordinate system. The orientation of the z″-axis with respect to the x′ y′ z′ coordinate system, and also with respect to the x y z coordinate system may therefore be assumed to be known. In this embodiment, calibration points721a,721b, . . . may be displayed simultaneously, as projections of screen marks521a,521b, . . . , and may differ in appearance, such that they may be distinguished by image processing software. Alternatively, calibration points721a,721b, . . . may appear as projections of screen marks521a,521b, . . . in an automatically controlled sequence. The user may then be queried todirect pointing device20 in the general direction ofinteraction region71 in such a way that the digital camera may image the calibration point under consideration. An automated calibration sequence may then be executed, wherein the z″-axis is directed, in sequence, tocalibration points721a,721b, . . . etc. The image processing software may identify thevarious calibration points721a,721b, . . . and the 3D position of these points may then be determined from knowledge of the orientation of the z″-axis and the coordinates of the z″=0 position with respect to the x y z coordinate system, in addition to the measured point-of-aim-distance211 between the calibration points721a,721b, . . . and the origin of the x′ y′ z′ coordinate system. This embodiment may also provide light-beam projection device202 mounted in a way that enables control of the direction of the light-beam relative to the x′ y′ z′ coordinate system. This light-beam may be used to aid in positioning the z″-axis. Embodiments where light-beam projection device202 are fixed or left out entirely are also contemplated.
• The present invention also contemplates situations in which parts of the calibration procedures outlined may be repeated at appropriate times. For example, when the relative position ofbase station30 andinteraction region71 changes substantially there may be a need to recalibrate the system. Such may also be the case when the position with respect tointeraction region71 of pointingdevice20 changes significantly, while operation of a particular embodiment of the invention operated using the assumption that such position remains substantially unchanged. As another example, coordinatesensing device201 and/or coordinatesensing device301 may include time-integrated acceleration measuring devices. In such a case accuracy of coordinate sensing may deteriorate over time, because of drift in the base acceleration measurements. When the deterioration has become large enough to be unacceptable to the user, the user may be queried to placepointing device20 in certain defined positions with respect to the x y z coordinate system, so as to re-initialize coordinatesensing device201 and301.
• The present invention also contemplates the use of a plurality of pointingdevices20. Each of the plurality of pointingdevices20 may be uniquely identifiable by the computer, for example by carrying a unique ID code inside their respective communication andcontrol device204. Each of the plurality of pointing devices may be associated with a specific cursor and/or aspecific interaction region71. For example, a first pointing device may be used to control a corresponding cursor on the left-hand one of twointeraction regions71, and a second pointing device may be used to control a corresponding cursor on the right-hand one of twointeraction regions71. Alternatively, both pointing devices may be used to control a corresponding, identifiable cursor on thesame interaction region71.
• Furthermore, the present invention contemplates the use of a plurality of entities such asbase station30, a particular one of which may be associated with the origin of the x y z coordinate system. Such a base station may be designated as the ‘master base station’. Each of the base stations may include coordinatesensing device201, so that their respective position and orientation relative to the master base station can be determined with greater accuracy than that afforded by the coordinatesensing device201 that is incorporated in pointingdevice20. Using such multiple base stations and coordinate sensing devices, 3D data pertaining to pointingdevice20 may be determined relative to the closest base station, and the relative position and orientation of that base station with respect to the master base station may be used to establish the 3D data pertaining to pointingdevice20 with respect to the x y z coordinate system. This may be of particular use in a situation where a user desires to move to a position in the room not within practical range of the master base station. In such a case, the user may take one of the additional base stations along. Since the position and orientation of this base station with respect to the master base station may be established with higher accuracy, this base station may be viewed as a relay-station, effectively increasing the operational range of the entire system. Note that, in such a case, the position of this base station may be changed with respect tointeraction structure72 without the need to re-execute the calibration procedures. Alternatively, signals from multiple base stations, as measured by the coordinatesensing device201 disposed in pointingdevice20, may be used simultaneously to reduce measurement errors. Generally speaking, determining the 3D data pertaining to pointingdevice20 with respect to the x y z coordinate system by making use of only onebase station30 may be less accurate than by making use of a plurality of base stations. Hence, the effective range of operation of pointingdevice20 may be enhanced by distributing a plurality of base stations over a large region of space.
• Referring toFIG. 17; in another embodiment, different devices than light-beam projection devices may be used to determine the position of the points in set P. For example, light-beam projection device202 may be omitted in an embodiment wherepointing device20 has the general shape of an elongated body, such as a pointing stick. The length of this stick may be predetermined or variable. For example, pointingdevice20 could be in the shape of an extendable, or telescopic pen such as the INFINITER Laser Baton sold by Bluesky Marketing—Unit 29, Six Harmony Row, Glasgow, G51 3BA, United Kingdom. The process of highlighting acalibration point721a,721b, . . . , in this embodiment may be replaced by the action of pointing to thecalibration point721a,721b, . . . , thereby guided by the elongated shape of pointingdevice20 instead of the light spot shown inFIG. 16 at point-of-aim210. Distance measuringdevice206 may also be provided in a related embodiment, similar to the second described embodiment. For example,distance measuring device206 may include a device to measure the distance from the tip of the elongated body of pointingdevice20 to the origin of the x′ y′ z′ coordinate system. In such an embodiment, in order to obtain measurements of point-of-aim-distance211, as described for example atprogram element130f(seeFIG. 12), the user might be queried to make physical contact between the tip of pointingdevice20 and the calibration point. Referring toFIG. 17, pointingdevice20 may also comprise contact-sensingdevice207, for example at the tip of pointingdevice20, capable of indicating physical contact between the tip of pointingdevice20 and a surface. Contact-sensingdevice207 may be in the form of a pressure-sensor or switch. Such sensors are known in the art.
• The present invention also contemplates the use of various other sensors integrated inpointing device20 and/orbase station30 to help determine the position and/or orientation ofinteraction structure72. For example, pointingdevice20 and/orbase station30 may include (ultra)sonic emitting and detecting devices to enable measuring substantially the shortest distance between a known point in the x y z coordinate system (e.g., the position of pointingdevice20 or of base station30) and the plane in whichprojection region60 lies. The user may be queried to alignpointing device20 orbase station30 such that the distance measuring device is oriented substantially perpendicularly toprojection region60. It should be understood that other types of sensors may be included in pointingdevice20 orbase station30 to help constrain the position and/or orientation ofinteraction structure72.
• It may be advantageous to put some parts disclosed in this application in a known relation to one another. For example,projection device40 andbase station30 may advantageously be incorporated in one and the same physical device, or be in a fixed relationship to one another. Alternatively,projection device40 may be equipped with a coordinatesensing device201. Whatever means are used, the system may be set up such that the position and/or orientation of some components, such as projection device40 (and its components) with respect to coordinate system x y z is fixed, known, or can be measured. If additional appropriate sensors are included (such as, for example,level sensing device303,visible markers302, etc.), even the position and/or orientation of moveable parts ofprojection device40 with respect to coordinate system x y z may be considered measurable and hence known. In particular, such an integrated device may furthermore incorporate a device like pointingdevice20, associated with apointing line21; that is, it may be equipped with a device that allows the orientation and/or position with respect to coordinate system x y z of apointing line21 to be known, measured and/or adjusted (potentially automatically). The preferred embodiment that results is depicted inFIG. 19 where, by example, the pointing line associated withintegrated pointing device20 has been made to coincide with theoptical axis401 ofprojection device40. In this application,optical axis401 is defined as the line substantially connecting theoptical origin402 of the projection beam (see alsoFIG. 21) with the projected center ofcomputer screen image50, as projected byprojection device40. (Note that the projected center ofcomputer screen image50 may coincide with the center ofprojection image70, notably so if both are rectangular.) Alternatively, the pointingline21 may substantially coincide with another line connecting a characteristic point of projected image70 (for example, one of its corners) to theoptical origin402 ofprojection device40, or with other convenient lines. An even more advantageous embodiment may result when thepointing device20 that is integrated withbase station30 andprojection device40 also incorporates distance measuring device/image capture device206, as indicated by the appearance of point-of-aim-distance211 inFIG. 19. To explain, it will now be assumed that the origin of the x y z system coincides with the origin of the x′ y′ z′ system of thepointing device20 that is integrated withbase station30 andprojection device40, as well as with theoptical origin402 of the projection beam ofprojection device40. It will also be assumed that the aspect ratio is known of a substantially rectangularcomputer screen image50 that is to be projected byprojection device40. If one furthermore assumes thatprojection device40 is aligned in such a way that itsoptical axis401 is substantially perpendicular to (a flat)projection region60 and that its optical characteristics (such as the dimensions of its projection beam and the size of the to-be-projected image) are known, it will be clear to those skilled in the art that knowledge of the distance between the optical origin ofprojection device40 and (the center of)projection image70 is sufficient to uniquely establish the position, shape, orientation and size of projection image70 (andinteraction region71 andinteraction structure72, assumed in this example to substantially coincide with projection image70), as measured with respect to the x y z coordinate system. Notably, in such a scenario there would not be a need for the use of a user-wieldedpointing device20 to aid in establishing the above parameters. Indeed, under these circumstances a user-wieldedpointing device20 may be used as a direct-pointing instrument straight-away, without user-assisted calibration, after the above described automated calibration procedure has been executed so as to establish all necessary parameters of projection image70 (andinteraction region71 and interaction structure72).
• If the integrated device (formed by at least combining, in the fashion described above,projection device40 andbase station30, and possibly a pointing device20) is not aligned perfectly perpendicular toprojection region60, an effect commonly referred to as keystoning will result in aprojection image70 that is not rectangular. For example, if theoptical axis401 is perpendicular to (a vertically oriented)projection region60 in a horizontal but not in a vertical sense, projection of a rectangularcomputer screen image50 will generally result in aprojection image70 that is an isosceles trapezoid with two horizontal edges. This is the most common form of image distortion, and is known as vertical keystoning. Conversely, if theoptical axis401 is perpendicular in a vertical, but not in a horizontal sense, the resulting isosceles trapezoid will have two vertical edges; this effect is known as horizontal keystoning. If both horizontal and
• vertical misalignments are not too severe, the assumption that projection image70 (andinteraction region71 and interaction structure72) is rectangular may still result in a system that affords the feeling of direct pointing, even if the system also assumes the vertical and horizontal alignments are perfectly perpendicular; the reasons for this are explained in paragraph95.
• In the more general case, as depicted in prior artFIG. 20, one or both of these misalignments may be substantial. In such a case, the keystone effect will result in aprojection image70 that is neither rectangular nor shaped like an isosceles trapezoid. As explained in numerous prior art (such as U.S. Pat. No. 6,520,647 to Mitsubishi Electric Research Laboratories Inc., and U.S. Pat. No. 6,877,863 to Silicon Optix Inc.), vertical keystoning can easily be corrected for by incorporating sensors likelevel sensing device303—see also prior artFIG. 20. Using such sensors, the vertical angle between the (vertically oriented)projection image70 andoptical axis401 can easily be established. If, furthermore, distance measuringdevice206 is subsequently used to substantially establish the distance between the projected version of the center ofcomputer screen image50 and theoptical origin402 of projection device40 (i.e., measured along the optical axis401), those skilled in the art will appreciate that inclusion of information on the optical characteristics ofprojection device40 will, once again, provide sufficient information to be able to establish position, shape, orientation and size of aninteraction structure72 that is near-coincident with aninteraction region71 andprojection image70, provided the horizontal keystone effect is not too severe. Note that this may be done independent of whether or not the vertical keystone effect is corrected for.
• If the horizontal keystone effect is also severe, the above cited prior art teaches that additional sensors may be used to establish the horizontal angle betweenprojection region60 andoptical axis401, and that they may be used to correct for the horizontal keystone effect. Using both vertical and horizontal keystone correction, a situation as depicted in prior artFIG. 20 may thus arise, characterized by a non-perpendicular alignment both in horizontal and vertical sense, nonetheless resulting in a substantially undistortedrectangular projection image70. Such sensors may be devices such as used in distance measuring device/image capturing device206—see also prior artFIG. 20. Alternatively,distance measuring device206 may be used to establish the distance from theoptical origin402 of the projection beam to two well-chosen characteristic points ofprojection image70. One characteristic point might be the projected center ofcomputer screen image50; the other one might be the projected version of a point half-way the left vertical edge ofcomputer screen image50. Simple application of the cosine rule (seeFIG. 21, depicting a top-view ofFIG. 19 for a case where the horizontal alignment is not perfectly perpendicular) shows that, if distances b and c and angle α are known (the latter angle following from knowledge of the optical dimensions of the projection beam), all other quantities in the triangle can be calculated—including the sought-after angle β. Given the above, it will be appreciated that inclusion of any such sensors will allow a unique determination of position, shape, orientation and size of projection image70 (and a near-coincident interaction region71 and interaction structure72), even in the general case of a projection device40 (integrated in the above described sense with at leastbase station30 and possibly a pointing device20) that is misaligned with respect toprojection region60 both in the horizontal and vertical sense. Note again that this may be done independent of whether or not the vertical and/or horizontal keystone effect is corrected for.
• Finally, then, it will be understood that use of animage capturing device206 of which the position and orientation in the x y z coordinate system is known may be of great help in the presently discussed embodiment, as already alluded to in paragraph158 and159. Indeed, those skilled in the art will recognize thatdistance measuring device206 may be embodied by combining knowledge of the optical characteristics of projection device40 (such as the dimensions of its projection beam and the size of the to-be-projected image) with information obtained from a captured image ofprojection image70 by a sensor likeimage capture device206. After all, if the size of projection image70 (for ease of explanation, but not out of necessity, assumed to be rectangular) can be established using animage capture device206, one may use knowledge of the optical characteristics ofprojection device40 to easily establish the distance from itsoptical origin402 to, for example, the projected center ofcomputer screen image50. This way, then, 3D parameters describingprojection image70,interaction region71 andinteraction structure72 can easily be determined, as will be appreciated by those skilled in the art. Alternatively, focus mechanisms inimage capture device206 may also be found useful in this regard. In short, it will be understood that any device capable of establishing the distance from a known point in the x y z coordinate system (such as, for example, theoptical origin402 of the projection beam of projection device40) to a point with a predetermined relation toprojection image70,interaction region71 orinteraction structure72 is contemplated by the current application. It will also be understood that, although the above continuously referred to the projected center ofcomputer screen image50, this was merely intended as an example and such a point may, without loss of functionality, be replaced by any point related in a known fashion tocomputer screen image50,projection image70,interaction region71 orinteraction structure72.
• Furthermore, other techniques disclosed in this application may be combined with the above described methods, including but not limited to those outlined in paragraphs149,155,160,161,162,163 and164. In particular, calibration sensors associated withprojection device40 orbase station30 or anintegrated pointing device20 and their associated calibration methods may be combined with methods related to a user-wieldedpointing device20, as disclosed in this application. For example, the system may require the user to highlight one or more calibration marks721a,b, . . . with a user-wieldedpointing device20, in order to correct for the horizontal non-alignment ofprojection device40 and projection region60 (the horizontal keystone-effect), while sensors incorporated inprojection device40 orbase station30 may be used to automatically correct for any vertical misalignment. Also, control points (see also paragraph147) may be used to check the validity of the calculated position, orientation, shape and size ofprojection image70,interaction region71 and/or a near-coincident interaction structure72. Also, it will be understood that in certainembodiments base station30 may be left out entirely. Furthermore,distance measuring device206 may also advantageously be left out. To explain, the reader is referred to the embodiment described in paragraphs starting at paragraph126. If the methods and devices described there are combined with the methods and devices outlined in paragraphs167-171, another advantageous embodiment will result. That is to say, if the position of a user-wieldedpointing device20 is not too different from theoptical origin402 of the projection beam ofprojection device40, such that they may be considered to substantially coincide, it will be appreciated by those skilled in the art that a system results that does not have a need for position sensing, an associatedbase station30 ordistance measuring device206. Particularly whenprojection region60 is sufficiently far away fromprojection device40, such thatprojection image70 is relatively large, any small discrepancy between the position of a user-wieldedpointing device20 and the position ofprojection device40 will be negligible. In such a case, knowledge of the dimensions of the projection beam (assuming that the angle between itsoptical axis401 andprojection region60 is substantially known) will be sufficient to substantially ascertain the point-of-aim of user-wieldedpointing device20, as will be obvious from the methods disclosed in this application. An improvement on the above system may be obtained by including one or moredistance sensing devices206 that may, for example, be used to establish the horizontal and/or vertical angles betweenoptical axis401 andprojection region60, as described above. This way, it is possible to compensate for a non-perpendicular alignment that might, without inclusion of such sensors, cause intolerable errors in the computation of the point-of-aim of a user-wieldedpointing device20.
• Yet another preferred embodiment results as a variation on a combination of the embodiments described in paragraphs120-125 and paragraphs126-145, and will be disclosed here with reference toFIGS. 3, 11, 12, 22, and 23. It will be understood that, for reasons of clarity,projection device40 has not been drawn inFIG. 23. The present embodiment is characterized by apointing device20 that is equipped with at least adistance measuring device206 and a coordinatesensing device201 that may be limited to sensing orientation. Such a coordinatesensing device201 may for example rely on an electronic compass like the HMR3300, on gyroscopes, on combinations of the above, or on any other suitable technology. It will be assumed that the position of pointingdevice20 does not change substantially while the calibration procedures described inFIGS. 3, 11 and 12 are executed, as indicated inFIG. 23 by the two near-coinciding instantiations of pointingdevices20. For purpose of explanation it will therefore be assumed that the origin of coordinate system x y z and the origin of coordinate system x′ y′ z′ are coincident, at least while the above-mentioned calibration procedures are executed. Finally, it will be assumed that one of the axes of the orthogonal coordinate system x y z is substantially vertical while the other two are substantially horizontal (the inclusion of a level-sensing device303, not specifically shown inFIG. 22 for reasons of clarity, may be beneficial in this context).
• The operation of the present embodiment will now be described with reference toFIGS. 3, 11, and 12. The process elements inFIG. 3 will be identical to the ones described in the first embodiment, resulting in similar assumptions regarding the shape of interaction region71 (and interaction structure72) and possibly orientation and position ofinteraction structure72. In particular, a default distance betweeninteraction structure72 and the origin of coordinate system x y z will be assumed for purpose of initialization.
• With reference toFIG. 11, program flow now continues at120a. Program elements120a-120hare shown to be similar to program elements90a-90m(FIG. 4), except that only point CA, set A and counter a are considered. That is to say, in the present embodiment it is no longer necessary to determine the second point CB and associated repositioning ofpointing device20 during the calibration procedure, since the position of pointingdevice20 is assumed not to change substantially from the origin of coordinate system x y z during execution of the steps considered.
• Afterprogram element120h, program flow continues with program elements outlined inFIG. 12. Referring toFIG. 12, program elements130a-130gare seen to be similar to steps100a-100pofFIG. 5. Comparingprogram element130fwithprogram element100f(FIG. 5), it can be seen that at130fthe program also stores information on the point-of-aim-distance between the origin of the x′ y′ z′ coordinate system and at least one of points C(p), indicated inFIG. 23 as721d. Comparingelements130gand100p(FIG. 5), it can be seen that these point-of-aim-distances (211 inFIG. 23) are also used in determining the 3D coordinates of the points in set P.
• With complete information on the 3D coordinates of the points in set P, it is then possible to construct a 3D description of interaction structure72 (seeFIG. 23). Therefore, program elements110a-110kas described earlier with reference toFIG. 6 may also be followed in the present embodiment.
• As in previously a described embodiment, the above outlined embodiment may be satisfactorily used as a direct-pointing system if the position of pointingdevice20 does not change substantially from the origin of coordinate system x y z. When the user moves to a different part of the room the position of pointingdevice20 will, by assumption, be unknown (since coordinatesensing device201 was assumed to only be capable of sensing orientation). This position then needs to be reestablished in order to be able to execute steps like110fand110h(seeFIG. 6), which allow the system to be used for direct-pointing cursor control.
• In order to reestablish the position of pointingdevice20 the user may be required to direct pointingline21 to substantially pass through a point of which the 3D coordinates are substantially known in coordinate system x y z. In doing so, the user may but need not be assisted by light-beam projection device202. This point could, for example, be a corner point or any other well-defined point of interaction structure72 (which, by assumption, coincides with interaction region71). Advantageously, it could be the last position that projected cursor701 (seeFIG. 23) was directed to. After all, the 3D characteristics of interaction structure72 (and interaction region71) are known, as is the position ofcomputer cursor501; the inverse of map M can then easily be used to compute the 3D characteristics of projectedcursor701. It will also be apparent that measurements of the orientation of pointingline21 and point-of-aim-distance211 will then constitute enough information to calculate the position of pointingdevice20 in coordinate system x y z. With the position of pointingdevice20 reestablished, the present embodiment can again successfully be used to control the computer cursor in a direct-pointing manner. This will remain to be the case as long as the position of pointingdevice20 does not stray far from the recently reestablished position.
• The applicant recognizes that the action of directing pointingline21 through the above-mentioned well-defined point (whilst measuring point-of-aim-distance211 and orientation of pointing device20) may result in apointing line21 not perfectly passing through this point, especially if the user is not aided by light-beam projection means202. Indeed, any errors made during this action will result in similar mismatches between subsequently calculated points-of-aim and true points-of-aim of pointingdevice20. These mismatches may not be too apparent to the user though since, presumably, the user was not capable of discerning such a mismatch when attempting to direct pointingline21 through the above mentioned well-defined point, as will be appreciated by those skilled in the art. Consequently, the present embodiment may still afford the user the feeling of direct-pointing cursor control under such circumstances.
• Sensors capable of sensing position may but need not be added to the coordinatesensing device201 of the current embodiment. Indeed, even if these sensors were only capable of maintaining positional accuracy over a limited period of time (e.g., sensors based on accelerometer technologies), their inclusion could still be advantageous. For example, such sensors may be employed to accurately track the position of pointingdevice20 over the relatively short period of time that starts when the user directs pointingline21 through the above-mentioned well-defined point (whilst measuring point-of-aim-distance211 and orientation of pointing device20), and ends when the user indicates his or her desire to stop direct-pointing cursor control. This way, direct-pointing cursor control can still be maintained even if the position of pointingdevice20 strays relatively far from the position it was in at the start of the above-mentioned period of time. In a similar way, such position sensors may also be beneficial during the (relatively short) calibration procedures aimed at establishing the 3D characteristics of interaction structure72 (and, by assumption, interaction region71), and can help to reduce errors that are due to a violation of the assumption that the position of pointingdevice20 remains constant. The inclusion of such sensors may even obviate the need fordistance measuring device206 during calibration, when they are used according to the methods of earlier described embodiments (see for example paragraphs107-113).
• Distance measuringdevice206 may be embodied by a laser sensor such as implemented in the TruLaser TLM100 Distance Measurer manufactured by Stanley. Such sensors are exceptionally well suited for the present embodiment, being capable of accurately determining distance along apointing line21 to a point on a plane, even if the angle between the plane's normal andpointing line21 is large. Other distance sensors that experience more difficulty under such circumstances may however also be used satisfactorily. For example, (ultra)sonic sensors such as the MaxSonar-EZl as manufactured by Maxbotix have a measurement beam that is far broader than that of the above mentioned laser-sensor. Such (ultra)sonic sensors are known to be susceptible to measurement errors when the beam's central axis is at a large angle to the normal of a smooth plane it is aimed at. Approximate knowledge of this angle may conceivably be used to at least partially correct for such errors, as will be appreciated by those skilled in the art.
• A system based on coordinatesensing device201 capable of sensing orientation (such as the HMR3300 electronic compass) in combination with an (ultra)sonic distance measuring device206 (such as the MaxSonar-EZl) constitutes a particularly economical and consequently advantageous preferred embodiment, which will now be described. Using methods outlined in the embodiment described in paragraphs126-145, the orientation, size and position of aninteraction structure72 that is a scaled and parallel version ofinteraction region71 can readily be found. For example, if it is assumed thatinteraction region71 is substantially a rectangle with horizontal and vertical sides, it is sufficient to consider only 3 calibration lines. If it is furthermore assumed that the aspect ratio of this rectangle is known, only 2 calibration lines suffice to narrow the number of possible solutions forinteraction structure72 down to 2. This can be understood by considering an example of two calibration lines passing through lower left corner C(3) and upper right corner C(2), depicted as721dand721ainFIG. 23. If the lower left corner of interaction structure72 (which is assumed to be a scaled and parallel version of the rectangular interaction region71) lies at such a distance from the origin that its z-coordinateequals 1, we find
• $\begin{array}{cc}C\left(2\right)={\mathrm{<figure-callout\; label="\lambda "\; filenames="US20190317613A1-20191017-D00004.png,US20190317613A1-20191017-D00005.png"\; state="\{\{state\}\}">\lambda </figure-callout>}}_2·\left(\begin{array}{c}\mathrm{Rx}2\\ \mathrm{Ry}2\\ Rz2\end{array}\right)& \left(21\right)\\ C\left(3\right)=\left(\begin{array}{c}\mathrm{LLx}\\ \mathrm{LLy}\\ 1\end{array}\right)& \left(22\right)\end{array}$
• Here, LLx, LLy, Rx2, Ry2 and Rz2 are known, since C(3) and C(2) lie on pointinglines21 of which the orientation has been measured. If we subsequently require these points to define a rectangle with horizontal and vertical sides and aspect ratio rat, we find
• 
  √{square root over ((LLx−λ₂·Rx2)²+(LLy−λ₂·Ry2)²)}=rat·√{square root over ((λ₂·Rz2−1)²)}⇒
• 
  (Ry2²+Rx2²−rat²·Rz2²)·λ₂²+
• 
  (−2·LLx·Rx2−2·LLy·Ry2+2·rat²≤Rz2)·λ₂+
• 
  (LLx²−rat²+LLy²)=0  (23)
• which can readily be solved for λ₂. Because of the quadratic nature there will in general be 2 solutions to Equation (23), and additional data is needed to pick the correct solution. This may be in the form of a third line passing through yet another characteristic point ofinteraction structure72, but it may also be in the form of knowledge of the (approximate) angle between the plane in whichinteraction structure72 lies and a line connecting the origin to one of its characteristic points. The applicant realizes that the above equation may not always generate real solutions for λ₂, since the pertinent determinant may be negative due to measurement errors, a mismatch between the true and assumed aspect ratio, etc. Those skilled in the art will appreciate, however, that there are many ways to find real values for λ₂that approximately solve Equation (23). For example, the value for rat may be slightly varied from its a-priori value to the point where Equation (23) yields real solutions, which may be acceptable as solutions to the original problem. Other ways of finding approximate solutions are also contemplated by the current application.
• It will be understood that aninteraction structure72 that substantially coincides withinteraction region71 can be readily found if the distance from the origin to at least one of the characteristic points of the latter is measured. Such a measurement can be made using accurate sensors based on (for example) laser technology. If (ultra)sonic sensors are used for this purpose, knowledge of the angle between the (ultra)sonic beam (for purpose of explanation assumed to be substantially centered on pointing line21) and the normal to the plane in whichinteraction region71 lies may prove beneficial, as will be appreciated by those skilled in the art of (ultra)sonic distance measurement. This angle can be readily calculated, as outlined above, once the orientations of a minimum of two calibration lines have been measured, and may advantageously be used during processing of any (ultra)sonic distance measurement(s).
• The accuracy and reliability of the (ultra)sonic distance measurement may be greatly enhanced if the angle between the (ultra)sonic beam and the normal to the plane in whichinteraction region71 lies is small. For this reason, the user may advantageously be asked to execute the entire calibration procedure from a position that meets such a requirement. It will be obvious that the fulfillment of this requirement may be tested once the orientation of at least 2 calibration lines has been measured. In general, the user may be directed to perform the required distance measurement(s) whilst orientingpointing device21 in one or more directions deemed most advantageous. For example, the user may be asked to substantially point at the center ofinteraction region71, at certain points on its edges, or trace out a diagonal ofinteraction region71 whilst multiple distance measurements are taken. Many other ways of measuring a suitable distance parameter are possible. With reference toFIG. 24 (for purpose ofclarity projection device40 has been omitted from the drawing, andinteraction region72 andinteraction structure71 have been depicted as coincident), it may in some cases be advantageous to take a distance measurement whilst directingpointing device20 at a point that does not lie ininteraction region71, but has a defined relationship to it. Such points may include apoint721ethat substantially lies in the same horizontal plane as pointing device20 (that is, with vertical coordinate substantially equal to 0) as well as in the plane in whichinteraction region71 lies (which is assumed vertical). This will result in a measurement for which the (ultra)sonic beam and the plane in whichinteraction region71 lies are substantially perpendicular in the vertical sense, which may constitute more optimal conditions for the use of (ultra)sonic measurement devices. This can be advantageous in situations whereinteraction region71 is mounted so high above pointingdevice20 that anypointing line21 pointed atinteraction region71 will have a large angle with the normal to the plane in whichinteraction region71 lies, as depicted inFIG. 24. Even more optimal circumstances might be created if point721 is chosen such that pointingline21 perpendicularly intersects the plane in whichinteraction region71 lies in both horizontal and vertical sense. Other points that have a defined relationship tointeraction region71 may also be used advantageously. It will be evident that the validity of desirable conditions (such as the (ultra)sonic beam being near-horizontal) can easily be checked, since the orientation of pointingdevice20 with respect to coordinate system x y z, as well as with respect tointeraction region71, are known, measurable or can be calculated.
• Even withoutdistance measuring device206, the outlined methods provide for a direct-pointing cursor control system as long as pointingdevice20 substantially remains at one and the same position. When the user decides to move to a different location in the room, depicted inFIG. 24 as the third instantiation of pointingdevice20, the position of pointingdevice20 will be unknown and has to be at least partially reestablished. As explained earlier, this may be accomplished by requiring the user to direct pointingline21 through a well-defined point of which the 3D coordinates are known in coordinate system x y z, whilst measuring the orientation of pointingline21 and point-of-aim-distance211. However, an (ultra)sonicdistance measuring device206 may not be capable of establishing point-of-aim-distance211 with sufficient accuracy, especially when the angle between pointingline21 andinteraction region71 is unfavorable. Also, measuring point-of-aim-distance211 may take too much time to be convenient during operation of the invention as a direct-pointing cursor control system. In such cases, with reference toFIG. 24, use may be made of the fact that an average user will tend to wieldpointing device20 in the same horizontal plane as from which the initial calibration routine was executed. Under such circumstances, the position of pointingdevice20 can be reestablished without the need for a distance measurement, namely as the intersection of pointingline21 with a horizontal plane with vertical coordinate z=0, i.e., with a plane through the origin of coordinate system x y z (chosen such that it contains the point from where the calibration procedures were executed). Advantageously, the well-defined point mentioned above may be chosen such that the angle between pointingline21 and the normal to the plane with vertical coordinate z=0 is as small as possible, so as to minimize the error in the calculated position. Other choices are also possible; good candidates may include points close to the upper edge ofinteraction region71 such as its upperright corner721b, depicted inFIG. 24. It will be appreciated that small errors in calculated position resulting from the fact that pointingdevice20 is not always held exactly at the same vertical position may still afford the user the feeling that the cursor is controlled in a direct-pointing manner. To the user, then, the invention affords a system where the cursor may be ‘grabbed out of the top right screen corner’, and subsequently moved about in a direct-pointing manner.
• Those skilled in the art will appreciate that the user may but need not be required to ‘grab the cursor’ every time he or she desires to engage in direct-pointing cursor control. For example, pointingdevice20 may be equipped with two buttons that activate cursor control, namely one that requires the user to firstdirect pointing line21 through a well-defined point and another that presumes the position of pointingdevice20 has not substantially changed since the last time the cursor was activated. Alternatively, the user may be required to indicate his or her desire to reestablish position in other ways, for example by executing a predefined motion pattern such as shakingpointing device20; coordinatesensing device201, possibly in combination with additional sensors such as accelerometers and gyros may be of help under these circumstances. Still alternatively, the system may include algorithms aimed at detecting when pointingdevice20 may have strayed far enough to warrant reestablishing its position in coordinate system x y z. Such detection algorithms may for example be based on elapsed time, on inertial measurements made by gyros and/or accelerometers that may be part of coordinatesensing device201, on distance measurement devices capable of sensing received power from a transmitting device fixed in space, etc. All of these and similar methods are contemplated by the invention.
• It will also be appreciated that inclusion of coordinatesensing devices201 that are capable of maintaining positional accuracy over a limited amount of time (e.g., inertial sensors) may obviate the need fordistance measuring device206 both during calibration as well as during operation of the present embodiment. This may be understood by considering calibration methods outlined in earlier described embodiments (see for example paragraphs107-113) m combination with methods outlined in paragraph184.
• Thus, methods and means are disclosed that afford a compact system for interacting with a presentation in a direct-pointing manner, whilst leaving the presenter free to move about the presentation venue. This system, moreover, obviates the need for a base-station30 that embodies a coordinate system x y z, and for a coordinatesensing device201 capable of continuously tracking positional coordinates of pointingdevice20 in this coordinate system.
• It will be obvious that elements of the various disclosed embodiments can advantageously be combined. In particular, combining the embodiments described in paragraphs167-172 and paragraphs173-188 will result in a system that may calibrate itself automatically, and leaves the presenter free to move about the presentation venue whilst engaging in direct-pointing interaction with the display. Moreover, the resulting system obviates the need for a base-station30 that embodies a coordinate system x y z, and for a coordinatesensing device201 capable of continuously tracking positional coordinates.
• In particular, when use is made ofdistance measurement devices206 for which the sensitive axis is relatively poorly defined (such as (ultra)sonic devices), methods disclosed in paragraphs167-172 and182-188 may still result in an economical and versatile system for which the calibration procedure is almost completely automatic. To clarify, reference is made toFIG. 25, whereprojection device40 is integrated with apointing device20 that contains at leastdistance measurement device206 and coordinatesensing device201. As explained in previous embodiments, this may result in a system capable of automatically establishing size, orientation and position of interaction structure72 (and, by assumption, interaction region71) as measured in a coordinate system x y z that has a predetermined relation toprojection device40. Moreover, since the line-of-sight betweenprojection device40 andinteraction region71 will in general be unimpeded, and since the angles pertinent to the problem are very well defined, those skilled in the art will appreciate. thatdistance sensing device206 may advantageously be embodied by (ultra)sonic sensors. What remains, then, is to find the vertical coordinate of the plane in which the user generally deploys the user-wieldedpointing device20, depicted inFIG. 25 as the ‘z=constant’ plane. To this end, the user-wieldedpointing device20 may also be equipped with adistance measuring device206, and the user may be required to perform a one-time calibration step by directing it towards a known point in coordinate system x y z whilst measuring both orientation of thepertinent pointing line21 as well as the pertinent point-of-aim-distance211. This known point can advantageously be chosen to coincide withprojection device40, particularly so ifdistance sensing devices206 of bothprojection device40 and user-wieldedpointing device20 are based on (ultra)sonic technology. In such a case it will be obvious to those skilled in the art that the pertinent point-of-aim-distance211 can be measured with relatively high accuracy, since (ultra)sonic transducers will be present both at the target and at the source. Alternatively, the user may be asked to direct user-wieldedpointing device20 to two or more known points in coordinate system x y z, such as top right and lower left corner ofinteraction region71, after which its position can be found by triangulation. In the latter case there is not even a need to incorporate adistance sensing device206 in user-wieldedpointing device20. Whatever choice is made, the above outlined methods yield enough information to establish the position of user-wieldedpointing device20 and, in particular, the z-coordinate of the plane in which it generally resides, so that methods outlined in paragraphs182-188 may be successfully employed to re-establish its position in case it is moved to a different location. As such, this system will be seen to be highly flexible, relatively inexpensive to manufacture, and extremely user-friendly. Indeed, the necessary equipment may come in the form of modules that can be retrofitted to existingprojection devices40 not specifically equipped for this application.
• It will be understood that in some cases it may be unnecessary to provide a separately embodiedbase station30, whilst still providing a coordinatesensing device201 capable of tracking orientational and positional coordinates. For example, coordinatesensing device201 may partly or completely rely on inertial sensing means such as accelerometers and gyroscopes, or on distance measurements with respect to walls, ceiling and/or floor. In such cases it would be possible to leave outbase station30 entirely and simply choose an appropriate x y z coordinate system. In such embodiments communication andcontrol device204 could be configured to communicate directly with the computer.
• The present invention also contemplates the use of automated procedures to establish the nature of the calibration points721a,721b,721c,721d, based on the measurements of the orientation and/or position of the pointing lines21 that substantially pass through them. For example, with reference toFIG. 7, calibration points721aand721dneed not be explicitly identified to the user as upper-right and lower-left corner respectively, but may merely be identified as diagonally opposite corners. Those skilled in the art will appreciate that in such a case an automated procedure may use the measurements of orientation and/or position of both pointinglines21, specifically with respect to the x-y-plane, to identify which of the two lines passes through the upper one of the two diagonally opposite corners. The practical consequence to the user is that the user need not be concerned about which of the twocalibration points721aand721dis highlighted first. Similar arguments can be used to construct automated procedures that determine whether apointing line21 substantially passes through a left or a right corner of interaction region71 (and, by assumption, of interaction structure72). For example, the a-priori assumption may be made that the clock-wise angle between a pointingline21 through a left corner and apointing line21 through a right corner, measured parallel to the x-y-plane, should not exceed 180 degrees, as will be appreciated by those skilled in the art.
• Finally, although the methods disclosed by the present invention are described in 3D, the invention may also be applied in 2D scenarios.
• While the invention has been described with respect to a limited number of embodiments, those skilled in the art, having benefit of this disclosure, will appreciate that other embodiments can be devised ‘which do not depart from the scope of the invention as disclosed herein. Accordingly, the scope of the invention should be limited only by the attached claims.

Claims (20)

What is claimed is:

1. A non-transitory computer-readable medium or media storing computer-executable instructions for directing a computer to perform a method for controlling the contents of a computer-generated image in conjunction with an enclosure comprising a plurality of sensors and configured to generate a pattern of light representing the image,

the plurality of sensors comprising:

an accelerometer;

a gyro;

an image capturing device;

a magnetometer; and

a sensing device that is sensitive to a position of the enclosure and orientation of the enclosure;

the method comprising the steps of:

receiving data that is dependent on an output of at least one of the plurality of sensors; and

controlling the contents of the computer-generated image based on the data.

2. The non-transitory computer-readable medium or media according toclaim 1, wherein the plurality of sensors further comprises a user input device configured to allow the user to manually provide two dimensional input, the user input device being insensitive to position and orientation of the enclosure.

3. The non-transitory computer-readable medium or media according toclaim 2, wherein the plurality of sensors further comprises a pressure sensor.

4. The non-transitory computer-readable medium or media according toclaim 2, wherein the pattern of light is generated by a display for displaying said image, the display being part of the enclosure.

5. The non-transitory computer-readable medium or media according toclaim 2, wherein the enclosure further comprises a laser configured to project a light spot in a direction that lies at a non-zero angle with the optical axis of the image capturing device.

6. The non-transitory computer-readable medium or media according toclaim 5, wherein

the laser is configured to project infrared light, and

the enclosure contains a computer-readable identification code that distinguishes the enclosure among a plurality of like enclosures.

7. The non-transitory computer-readable medium or media according toclaim 1, wherein the sensing device comprises the accelerometer, the gyro, the image capturing device and the magnetometer.

8. An apparatus for generating first data and second data for use in controlling a computer-generated image, the first data being dependent on a position of a first device relative to a second device, the second data being dependent on a position of the second device, the first device configured to be handheld and wielded in mid-air by a user, the second device configured to be taken along by the user, the apparatus comprising:

a first sensing device for generating the first data, the first sensing device being sensitive to a position of the first device relative to the second device; and

a second sensing device for generating the second data, the second sensing device being sensitive to a posit ion of the second device;

wherein the first device contains at least one element of the first sensing device.

9. The apparatus according toclaim 8, wherein the first data is dependent on an orientation of the first device, and the second data is dependent on an orientation of the second device.

10. The apparatus according toclaim 9, wherein

the first sensing device is sensitive to three independent positional coordinates of the first device relative to the second device and three independent orientational coordinates of the first device, and

the second sensing device comprises a magnetometer and is sensitive to three independent positional coordinates of the second device and three independent orientational coordinates of the second device.

11. The apparatus according toclaim 10, wherein

the first sensing device comprises a first accelerometer and a first gyro, and

the second sensing device comprises a second accelerometer and a second gyro and is sensitive to three independent positional coordinates of the second device relative to a point external to the first device and to the second device.

12. The apparatus according toclaim 11, wherein the first sensing device comprises a digital camera.

13. A non-transitory computer-readable medium or media storing computer-executable instructions for directing a computer to perform a method for controlling the contents of a computer-generated image in conjunction with an enclosure that contains a plurality of sensors and that contains a projection device,

the plurality of sensors comprising

an accelerometer and

a gyro and

a magnetometer and

an image capturing device and

a sensing device that is sensitive to a position of the enclosure and orientation of the enclosure;

the method comprising the steps of:

causing the projection device to project a plurality of distinct points,

causing the image capturing device to capture a first image,

receiving data that is dependent on an output of at least one of the plurality of sensors, and

using the data to control the contents of the computer-generated image when the data depends on at least a part of the first image and the first image contains a second image of at least one of the plurality of distinct points.

14. The non-transitory computer-readable medium or media according toclaim 13, wherein each of the plurality of distinct points is projected substantially simultaneously with the plurality of distinct points.

15. The non-transitory computer-readable medium or media according toclaim 13, wherein each of the plurality of distinct points is projected in a direction that has a non-zero angle relative to the optical axis of the image capturing device.

16. The non-transitory computer-readable medium or media according toclaim 15, wherein the projection device comprises a laser.

17. The non-transitory computer-readable medium or media according toclaim 16, wherein

the laser is for projecting infrared light, and

the enclosure contains a computer-readable identification code that distinguishes the enclosure among a plurality of like enclosures.

18. The non-transitory computer-readable medium or media according toclaim 13, wherein

the enclosure contains a computer-readable identification code that distinguishes the enclosure among a plurality of like enclosures, and

the plurality of sensors comprises a pressure sensor.

19. The non-transitory computer-readable medium or media according toclaim 13, wherein the sensing device is sensitive to two independent positional coordinates of the enclosure and three independent orientational coordinates of the enclosure.

20. The non-transitory computer-readable medium or media according toclaim 13, wherein the sensing device is sensitive to three independent positional coordinates of the enclosure and three independent orientational coordinates of the enclosure.

Images