Summary of the invention
For overcoming Problems existing in correlation technique, the disclosure provides a kind of information labeling method and device.
According to the first aspect of disclosure embodiment, a kind of information labeling method is provided, comprises:
According to each frame picture of multimedia file, obtain the scenario information of this multimedia file, this scenario information at least comprises lines information and corresponding temporal information;
According to the temporal information in the scenario information of this multimedia file, obtain the start time information of multiple multi-media segment in this multimedia file;
According to the start time information of the plurality of multi-media segment and the scenario information of this multimedia file, obtain the key message of the plurality of multi-media segment, each key message is for describing the summarized content of corresponding multi-media segment;
Set up the mapping relations between the key message of each multi-media segment in the plurality of multi-media segment and the start time information of corresponding multi-media segment.
In the first possibility implementation of first aspect of the present disclosure,
This is according to each frame picture of multimedia file, and the scenario information obtaining this multimedia file comprises:.
According to each frame picture of this multimedia file, by optical character identification OCR method, obtain the lines information in each frame picture of this multimedia file;
According to the reproduction time of each frame picture of this multimedia file, determine the temporal information that in each the frame picture with this multimedia file, lines information is corresponding;
According to lines information and corresponding temporal information in each frame picture of this multimedia file, determine the scenario information of this multimedia file.
In the second possibility implementation of first aspect of the present disclosure,
According to the temporal information in the scenario information of this multimedia file, the start time information obtaining multiple multi-media segment in this multimedia file comprises:
According to the temporal information in the scenario information of this multimedia file and predetermined interval duration, obtain the start time information of multiple multi-media segment in this multimedia file; Or,
Semantic analysis is carried out to the scenario information of this multimedia file, obtain multiple story of a play or opera information, according to the plurality of story of a play or opera information, this multimedia file is divided into multiple multi-media segment, a multi-media segment corresponds to a story of a play or opera, according to the temporal information of the plurality of multi-media segment with this multimedia file, obtain the start time information of the plurality of multi-media segment.
In the third possibility implementation of first aspect of the present disclosure, according to the start time information of the plurality of multi-media segment and the scenario information of this multimedia file, the key message obtaining the plurality of multi-media segment comprises:
For each multi-media segment, according to each frame picture of this multi-media segment and the lines information of this each frame picture, by face recognition technology, obtain the Role Information corresponding with the lines information of this each frame picture;
The Role Information corresponding according to the lines information with this each frame picture and the lines information of this each frame picture, generate the key message of this multi-media segment.
In the 4th kind of possibility implementation of first aspect of the present disclosure, should for each multi-media segment, according to each frame picture of this multi-media segment and the lines information of this each frame picture, by face recognition technology, obtain the Role Information corresponding with the lines information of this each frame picture and comprise:
Set up the character information database of this multimedia file, this character information database at least comprises Role Information in this multimedia file and face characteristic information corresponding to each role;
Each lines information corresponding according to this multi-media segment, obtains the source picture with this each lines information from this multi-media segment;
Obtain the face characteristic information in this source picture;
According to the face characteristic information in this source picture, from this character information database, obtain the Role Information that each lines information is corresponding with this.
In the 5th kind of possibility implementation of first aspect of the present disclosure, the key message of this multi-media segment also comprises the key picture of this multi-media segment.
In the 6th kind of possibility implementation of first aspect of the present disclosure, the method also comprises:
According to the start time information of this multi-media segment, obtain the labeling position in playing progress bar;
When the trigger action to labeling position in this playing progress bar being detected, show the key message of corresponding multi-media segment.
According to the second aspect of disclosure embodiment, a kind of information labeling device is provided, comprises:
Scenario information acquisition module, for each the frame picture according to multimedia file, obtains the scenario information of this multimedia file, and this scenario information at least comprises lines information and corresponding temporal information;
Start time information acquisition module, for according to the temporal information in the scenario information of this multimedia file, obtains the start time information of multiple multi-media segment in this multimedia file;
Key message acquisition module, for according to the start time information of the plurality of multi-media segment and the scenario information of this multimedia file, obtain the key message of the plurality of multi-media segment, each key message is for describing the summarized content of corresponding multi-media segment;
Mapping relations set up module, for setting up the mapping relations between the key message of each multi-media segment in the plurality of multi-media segment and the start time information of corresponding multi-media segment.
In the first possibility implementation of second aspect of the present disclosure, this scenario information acquisition module is used for:
According to each frame picture of this multimedia file, by optical character identification OCR method, obtain the lines information in each frame picture of this multimedia file;
According to the reproduction time of each frame picture of this multimedia file, determine the temporal information that in each the frame picture with this multimedia file, lines information is corresponding;
According to lines information and corresponding temporal information in each frame picture of this multimedia file, determine the scenario information of this multimedia file.
In the second possibility implementation of second aspect of the present disclosure, this start time information acquisition module is used for:
According to the temporal information in the scenario information of this multimedia file and predetermined interval duration, obtain the start time information of multiple multi-media segment in this multimedia file; Or,
Semantic analysis is carried out to the scenario information of this multimedia file, obtain multiple story of a play or opera information, according to the plurality of story of a play or opera information, this multimedia file is divided into multiple multi-media segment, a multi-media segment corresponds to a story of a play or opera, according to the temporal information of the plurality of multi-media segment with this multimedia file, obtain the start time information of the plurality of multi-media segment.
In the third possibility implementation of second aspect of the present disclosure, this key message acquisition module is used for:
For each multi-media segment, according to each frame picture of this multi-media segment and the lines information of this each frame picture, by face recognition technology, obtain the Role Information corresponding with the lines information of this each frame picture;
The Role Information corresponding according to the lines information with this each frame picture and the lines information of this each frame picture, generate the key message of this multi-media segment.
In the 4th kind of possibility implementation of second aspect of the present disclosure, this key message acquisition module is used for:
Set up the character information database of this multimedia file, this character information database at least comprises Role Information in this multimedia file and face characteristic information corresponding to each role;
Each lines information corresponding according to this multi-media segment, obtains the source picture with this each lines information from this multi-media segment;
Obtain the face characteristic information in this source picture;
According to the face characteristic information in this source picture, from this character information database, obtain the Role Information that each lines information is corresponding with this.
In the 5th kind of possibility implementation of second aspect of the present disclosure, the key message of this multi-media segment also comprises the key picture of this multi-media segment.
In the 6th kind of possibility implementation of second aspect of the present disclosure, this device also comprises:
Labeling position acquisition module, for the start time information according to this multi-media segment, obtains the labeling position in playing progress bar;
Display module, for when the trigger action to labeling position in this playing progress bar being detected, shows the key message of corresponding multi-media segment.
The third aspect, additionally provides a kind of information labeling device, comprising:
Processor;
For the storer of the executable instruction of storage of processor;
Wherein, this processor is configured to:
According to each frame picture of multimedia file, obtain the scenario information of this multimedia file, this scenario information at least comprises lines information and corresponding temporal information;
According to the temporal information in the scenario information of this multimedia file, obtain the start time information of multiple multi-media segment in this multimedia file;
According to the start time information of the plurality of multi-media segment and the scenario information of this multimedia file, obtain the key message of the plurality of multi-media segment, each key message is for describing the summarized content of corresponding multi-media segment;
Set up the mapping relations between the key message of each multi-media segment in the plurality of multi-media segment and the start time information of corresponding multi-media segment.
The beneficial effect that the technical scheme that disclosure embodiment provides is brought is:
The disclosure is by obtaining the scenario information of multimedia file, and according to the lines information in this scenario information, this multimedia file is divided into multiple multi-media segment, and then according to the temporal information in the plurality of multi-media segment and this scenario information, labeling position in Lookup protocol multimedia file playing progress bar, by obtaining the summarized content of corresponding multi-media segment according to semantic analysis, i.e. key message, reduce the waste of human resources, the batch processing to multimedia file can be realized simultaneously.
Should be understood that, it is only exemplary and explanatory that above general description and details hereinafter describe, and can not limit the disclosure.
Embodiment
For making object of the present disclosure, technical scheme and advantage clearly, below in conjunction with accompanying drawing, disclosure embodiment is described in further detail.
Here will be described exemplary embodiment in detail, its sample table shows in the accompanying drawings.When description below relates to accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawing represents same or analogous key element.Embodiment described in following exemplary embodiment does not represent all embodiments consistent with the disclosure.On the contrary, they only with as in appended claims describe in detail, the example of apparatus and method that aspects more of the present disclosure are consistent.
Fig. 1 is the process flow diagram of a kind of information labeling method according to an exemplary embodiment, as shown in Figure 1, comprises the following steps.
In a step 101, according to each frame picture of multimedia file, obtain the scenario information of this multimedia file, this scenario information at least comprises lines information and corresponding temporal information.
In a step 102, according to the temporal information in the scenario information of this multimedia file, obtain the start time information of multiple multi-media segment in this multimedia file.
In step 103, according to the start time information of the plurality of multi-media segment and the scenario information of this multimedia file, obtain the key message of the plurality of multi-media segment, each key message is for describing the summarized content of corresponding multi-media segment.
At step 104, the mapping relations between the key message of each multi-media segment in the plurality of multi-media segment and the start time information of corresponding multi-media segment are set up.
The method that disclosure embodiment provides, by obtaining the scenario information of multimedia file, and according to the lines information in this scenario information, this multimedia file is divided into multiple multi-media segment, and then according to the temporal information in the plurality of multi-media segment and this scenario information, labeling position in Lookup protocol multimedia file playing progress bar, by obtaining the summarized content of corresponding multi-media segment according to semantic analysis, i.e. key message, reduce the waste of human resources, the batch processing to multimedia file can be realized simultaneously.
In the first possibility implementation of the present disclosure, this is according to each frame picture of multimedia file, and the scenario information obtaining this multimedia file comprises:.
According to each frame picture of this multimedia file, by optical character identification OCR method, obtain the lines information in each frame picture of this multimedia file;
According to the reproduction time of each frame picture of this multimedia file, determine the temporal information that in each the frame picture with this multimedia file, lines information is corresponding;
According to lines information and corresponding temporal information in each frame picture of this multimedia file, determine the scenario information of this multimedia file.
In the second possibility implementation of the present disclosure, according to the temporal information in the scenario information of this multimedia file, the start time information obtaining multiple multi-media segment in this multimedia file comprises:
According to the temporal information in the scenario information of this multimedia file and predetermined interval duration, obtain the start time information of multiple multi-media segment in this multimedia file; Or,
Semantic analysis is carried out to the scenario information of this multimedia file, obtain multiple story of a play or opera information, according to the plurality of story of a play or opera information, this multimedia file is divided into multiple multi-media segment, a multi-media segment corresponds to a story of a play or opera, according to the temporal information of the plurality of multi-media segment with this multimedia file, obtain the start time information of the plurality of multi-media segment.
In the third possibility implementation of the present disclosure, according to the start time information of the plurality of multi-media segment and the scenario information of this multimedia file, the key message obtaining the plurality of multi-media segment comprises:
For each multi-media segment, according to each frame picture of this multi-media segment and the lines information of this each frame picture, by face recognition technology, obtain the Role Information corresponding with the lines information of this each frame picture;
The Role Information corresponding according to the lines information with this each frame picture and the lines information of this each frame picture, generate the key message of this multi-media segment.
In the 4th kind of possibility implementation of the present disclosure, should for each multi-media segment, according to each frame picture of this multi-media segment and the lines information of this each frame picture, by face recognition technology, obtain the Role Information corresponding with the lines information of this each frame picture and comprise:
Set up the character information database of this multimedia file, this character information database at least comprises Role Information in this multimedia file and face characteristic information corresponding to each role;
Each lines information corresponding according to this multi-media segment, obtains the source picture with this each lines information from this multi-media segment;
Obtain the face characteristic information in this source picture;
According to the face characteristic information in this source picture, from this character information database, obtain the Role Information that each lines information is corresponding with this.
In the 5th kind of possibility implementation of the present disclosure, the key message of this multi-media segment also comprises the key picture of this multi-media segment.
In the 6th kind of possibility implementation of the present disclosure, the method also comprises:
According to the start time information of this multi-media segment, obtain the labeling position in playing progress bar;
When the trigger action to labeling position in this playing progress bar being detected, show the key message of corresponding multi-media segment.
Above-mentioned all alternatives, can adopt and combine arbitrarily formation embodiment of the present disclosure, this is no longer going to repeat them.
Fig. 2 is the process flow diagram of a kind of information labeling method according to an exemplary embodiment.The executive agent of this embodiment can be server, and with reference to Fig. 2, this embodiment specifically comprises:
In step 201, according to each frame picture of multimedia file, obtain the scenario information of this multimedia file, this scenario information at least comprises lines information and corresponding temporal information.
In the disclosed embodiments, this multimedia file refers to video file, the method obtaining the scenario information of this multimedia file can be: according to each frame picture of this multimedia file, by OCR (OpticalCharacterRecognition, optical character identification) method, obtain the lines information in each frame picture of this multimedia file; According to the reproduction time of each frame picture of this multimedia file, determine the temporal information that in each the frame picture with this multimedia file, lines information is corresponding; According to lines information and corresponding temporal information in each frame picture of this multimedia file, determine the scenario information of this multimedia file.
Wherein, OCR method is used for picture recognition to become word, because the lines information in multimedia file exists with graphic form, therefore the lines information in each frame picture of multimedia file can be obtained by OCR method, concrete grammar can be: the character features extracting each word in picture, according to this character features, corresponding word is obtained from Characteristic Contrast database, this comparison database is for storing character set and corresponding character features collection, it should be noted that, the method extracting character features from picture is identical with the extracting method of character features in this comparison database, disclosure embodiment is not construed as limiting the character features used in OCR method and characteristic of correspondence extracting method.
Obtain except the scenario information of multimedia file except adopting said method, because multimedia file content is different, corresponding lines information also there are differences, therefore can also be different according to the content of multimedia file, adopts diverse ways to obtain the scenario information of multimedia file.
Such as, when this multimedia file is the video file of singing contest, only can obtain the lines information of assigned role in this multimedia file, this assigned role can be host, the temporal information corresponding according to the lines acquisition of information of this assigned role, to generate the lines information of this multimedia file.
Different according to multimedia file content, adopt diverse ways to obtain scenario information, the operand of server can be reduced to a certain extent.
In another embodiment of the present disclosure, if the shooting side of this multimedia file provides the scenario information of this multimedia file, under normal circumstances, this scenario information does not comprise temporal information corresponding to every lines, can directly according to lines information, obtain the source picture in the multimedia file corresponding with this lines information, and temporal information corresponding for this source picture is defined as the temporal information corresponding with this lines information, and then generate the scenario information of this multimedia file comprising lines information and corresponding temporal information.
Except adopting said method to obtain except the scenario information of multimedia file, additive method can also be adopted to obtain the scenario information of multimedia file, and disclosure embodiment is not construed as limiting this.
In step 202., according to the temporal information in the scenario information of this multimedia file, obtain the start time information of multiple multi-media segment in this multimedia file.
In this multimedia file, the start time information of multiple multi-media segment is for determining the labeling position of playing progress bar in this multimedia process, make user can pass through to perform play operation to this labeling position, watch corresponding multi-media segment, disclosure embodiment is not construed as limiting this.In this multimedia file, the start time information acquisition methods of multiple multi-media segment can be following two kinds:
The first, according to the temporal information in the scenario information of this multimedia file and predetermined interval duration, obtain the start time information of multiple multi-media segment in this multimedia file.
Temporal information in the scenario information of this multimedia file at least comprises the duration of this multimedia file and time corresponding to each lines information.Such as, when the duration of this multimedia file is 100 minutes, predetermined interval duration can be set to 20 minutes, so, in this multimedia file playing process, 4 labeling positions in playing progress bar, can be had, namely be respectively corresponding play time the position of 20,40,60,80 minutes; This predetermined interval duration also can be set to other and be greater than zero and any numerical value being less than media file duration, and disclosure embodiment does not do concrete restriction to this.
In addition to the above methods, because multimedia file content is different, the frequency that in multimedia file, climax fragment occurs is also different, therefore can also according to this predetermined interval duration of the curriculum offering of multimedia file, such as, when the content of multimedia file is groups of stars' concert, because every song duration is about 4 minutes, therefore this predetermined interval duration can be set to 4 minutes; According to this predetermined interval duration of multimedia file curriculum offering, multiple climax fragment can be divided out respectively, thus make user be easier to get its content of multimedia wanting to understand.
According to the multi-media segment obtained by said method, obtain the start time information of corresponding each multi-media segment.
The second, semantic analysis is carried out to the scenario information of this multimedia file, obtain multiple story of a play or opera information, according to the plurality of story of a play or opera information, this multimedia file is divided into multiple multi-media segment, a multi-media segment corresponds to a story of a play or opera, according to the temporal information of the plurality of multi-media segment with this multimedia file, obtain the start time information of the plurality of multi-media segment.
Wherein, semantic analysis is the one in natural language processing method, particularly, refers in the analysis words implication of pending text message and the basis of syntactic structure, obtains the summarized content that can represent this pending text message major significance.
According to the scenario information of multimedia file, the method by semantic analysis this multimedia file being divided into multiple multi-media segment can be: the word segmentation result obtaining lines information in scenario information, and part of speech and word sense tagging are carried out to this word segmentation result, scenario information corresponding for the adjacent lines information with the similar meaning of a word is divided into one section, and obtain the summarized content of this section of scenario information, then according to the scenario information after segmentation, this multimedia file is divided into multiple multi-media segment; Also other semantic analysis can be adopted to divide this multimedia file, and disclosure embodiment is not construed as limiting this.
Except the method by semantic analysis divides except multimedia file, multimedia file can also be divided by additive method, thus obtain multiple multi-media segment.Such as, whether can change according to each frame picture decision event scene of multimedia file, any frame picture corresponding when being changed by event scenarios is as waypoint, also instant, can using the end time of time corresponding for this frame picture as a upper multi-media segment and the start time of next multi-media segment.
Except employing said method carries out except this multimedia file division, additive method can also be adopted to divide, disclosure embodiment is not construed as limiting this, then according to by dividing the multi-media segment that multimedia file obtains, obtains the start time information of corresponding each multi-media segment.
In another embodiment of the present disclosure, if the shooting side of this multimedia file provides the scenario information of this multimedia file, under normal circumstances, this scenario information does not comprise temporal information corresponding to every lines, but this scenario information divides this multimedia file according to the story of a play or opera, can directly by the method according to the corresponding temporal information of lines acquisition of information in step 201, obtain the temporal information that every lines information is corresponding, again according to the multimedia file division result that scenario information provides, obtain multiple multi-media segment, determine first the lines information that each multi-media segment is corresponding, and then according to first lines information corresponding to this each multi-media segment and corresponding temporal information, determine the start time information of each multi-media segment.
In step 203, according to the start time information of the plurality of multi-media segment and the scenario information of this multimedia file, obtain the key message of the plurality of multi-media segment, each key message is for describing the summarized content of corresponding multi-media segment.
The key message of the plurality of multi-media segment can obtain according to the semantic analysis in step 202, repeats no more herein.
In another embodiment of the present disclosure, the key message of this multi-media segment can also comprise the Role Information of this multi-media segment, this Role Information at least comprises the role name of corresponding role, this role can be the role that the frequency of occurrences is the highest, also can be the role the highest with the key message correlativity of this multi-media segment, correspondingly, determine that the method for this role can be following two kinds:
The first, when this role is the highest role of the frequency of occurrences, according to each frame picture of this multi-media segment, pass through face identification method, obtain all roles that this each frame picture comprises, the occurrence number of statistics each role in this multi-media segment, is defined as Role Information included in the key message of this multi-media segment by the Role Information of maximum for occurrence number role.
Certainly, also can determine by additive method the role that this frequency of occurrences is the highest, disclosure embodiment is not construed as limiting this.
The second, when this role is with the role that the key message correlativity of corresponding multi-media segment is the highest, according to the crucial words in the key message of multi-media segment, obtain and specify lines information, this appointment lines information refers to and comprises the lines information that crucial words with this has the words of similar semantic, lines information and role corresponding to every lines information is specified according to these, add up each role to the number should specifying lines information, the Role Information of roles maximum for number is defined as Role Information included in the key message of this multi-media segment.
Certainly, also can determine this role the highest with the key message correlativity of corresponding multi-media segment by additive method, disclosure embodiment is not construed as limiting this.
It should be noted that, Role Information included in the key message of multi-media segment can be multiple, concrete number can be determined according to the content of key message, also can the role number the highest according to the frequency of occurrences determine, such as, when role's number that the frequency of occurrences is the highest is 2, Role Information corresponding for these two roles can be defined as Role Information included in corresponding multi-media segment; Certainly, can also be determined the number of this Role Information by additive method, disclosure embodiment is not construed as limiting this.
In the disclosed embodiments, the method comprising the key message of Role Information obtaining each multi-media segment can be: for each multi-media segment, according to each frame picture of this multi-media segment and the lines information of this each frame picture, by face recognition technology, obtain the Role Information corresponding with the lines information of this each frame picture; The Role Information corresponding according to the lines information with this each frame picture and the lines information of this each frame picture, generate the key message of this multi-media segment.
Wherein, face recognition technology is a kind of computer technology of carrying out identity verify according to face characteristic, and in the disclosed embodiments, the method extracting Role Information in picture can be:
Set up the character information database of this multimedia file, this character information database at least comprises Role Information in this multimedia file and face characteristic information corresponding to each role; Each lines information corresponding according to this multi-media segment, obtains the source picture with this each lines information from this multi-media segment; Obtain the face characteristic information in this source picture; According to the face characteristic information in this source picture, from this character information database, obtain the Role Information that each lines information is corresponding with this.
In another embodiment of the present disclosure, the method setting up the character information database of this multimedia file can be: the Role Information in this multimedia file can also comprise the Real Name of the corresponding performer of each Role Information, using the Real Name of corresponding for this each Role Information performer as index, from network, search package is containing the picture of corresponding performer's face, and the face characteristic information extracted in this picture, by this face characteristic information and this picture corresponding stored in the character information database of this multimedia file.
The character information database setting up this multimedia file can adopt said method, can adopt additive method, and disclosure embodiment is not construed as limiting this.It should be noted that, the face feature extraction method used when obtaining character information database is identical with the method for the face characteristic information in each the frame picture extracting this multi-media segment, and the face characteristic information comprised in this multimedia information data storehouse is identical with the face characteristic information extracted from each frame picture of this multi-media segment.
In another embodiment of the present disclosure, the key message of this multi-media segment also comprises the key picture of this multi-media segment.This key picture can be picture corresponding to the initial time of this multi-media segment, and also can be any frame picture in this multi-media segment, disclosure embodiment be not construed as limiting this.
In step 204, the mapping relations between the key message of each multi-media segment in the plurality of multi-media segment and the start time information of corresponding multi-media segment are set up.
By setting up the mapping relations between the key message of this multi-media segment and the start time information of this multi-media segment, make, when the cursor of mouse is moved to corresponding labeling position by user, this key message to be shown near this labeling position.
The method for building up of the mapping relations between the key message of this multi-media segment and the start time information of this multi-media segment can be: by the start time information corresponding stored of the key message of this multi-media segment and this multi-media segment to appointed area, this appointed area is for storing the initial time of multi-media segment and corresponding multi-media segment, and the initial time of this multi-media segment that is to say the labeling position in corresponding multimedia file playing progress bar; Certainly, also can set up corresponding relation between the key message of this multi-media segment and the start time information of this multi-media segment by additive method, disclosure embodiment is not construed as limiting this.
In step 205, according to the start time information of this multi-media segment, obtaining the labeling position in playing progress bar, when the trigger action to labeling position in this playing progress bar being detected, showing the key message of corresponding multi-media segment.
Labeling position in playing progress bar can highlight with small circle, makes user can find this labeling position fast, and can also arrange no shape according to marked content difference and highlight labeling position, disclosure embodiment is not construed as limiting this.
This trigger action can be long touch operation, also can be the operation that the cursor of mouse stops for a long time, particularly, when detect preset duration is greater than to the touch duration of labeling position or cursor dwell duration time, show the key message of corresponding multi-media segment at this labeling position.
It should be noted that this key message at least comprises the summarized content of corresponding multi-media segment, Role Information and key picture can also be comprised, certainly, can also comprise the information that other can give prominence to corresponding multi-media segment content, disclosure embodiment is not construed as limiting this.
The method that disclosure embodiment provides, by obtaining the scenario information of multimedia file, and according to the lines information in this scenario information, this multimedia file is divided into multiple multi-media segment, and then according to the temporal information in the plurality of multi-media segment and this scenario information, labeling position in Lookup protocol multimedia file playing progress bar, by obtaining the summarized content of corresponding multi-media segment according to semantic analysis, i.e. key message, reduce the waste of human resources, the batch processing to multimedia file can be realized simultaneously; Further, by adding Role Information and key picture in key message, make when the trigger action to labeling position being detected, the summarized content of corresponding multi-media segment can not only be shown, the Role Information in this corresponding multi-media segment and key picture can also be shown, thus make user more fully can understand content in corresponding multi-media segment.
Fig. 3 is a kind of information labeling device block diagram according to an exemplary embodiment.With reference to Fig. 3, this device comprises scenario information acquisition module 301, start time information acquisition module 302, key message acquisition module 303, and mapping relations set up module 304.
Scenario information acquisition module 301, for each the frame picture according to multimedia file, obtains the scenario information of this multimedia file, and this scenario information at least comprises lines information and corresponding temporal information;
Start time information acquisition module 302, for according to the temporal information in the scenario information of this multimedia file, obtains the start time information of multiple multi-media segment in this multimedia file;
Key message acquisition module 303, for according to the start time information of the plurality of multi-media segment and the scenario information of this multimedia file, obtain the key message of the plurality of multi-media segment, each key message is for describing the summarized content of corresponding multi-media segment;
Mapping relations set up module 304, for setting up the mapping relations between the key message of each multi-media segment in the plurality of multi-media segment and the start time information of corresponding multi-media segment.
The first providing in the disclosure may in implementation, this scenario information acquisition module 301 for:
According to each frame picture of this multimedia file, by optical character identification OCR method, obtain the lines information in each frame picture of this multimedia file;
According to the reproduction time of each frame picture of this multimedia file, determine the temporal information that in each the frame picture with this multimedia file, lines information is corresponding;
According to lines information and corresponding temporal information in each frame picture of this multimedia file, determine the scenario information of this multimedia file.
The second provided in the disclosure may in implementation, this start time information acquisition module 302 for:
According to the temporal information in the scenario information of this multimedia file and predetermined interval duration, obtain the start time information of multiple multi-media segment in this multimedia file; Or,
Semantic analysis is carried out to the scenario information of this multimedia file, obtain multiple story of a play or opera information, according to the plurality of story of a play or opera information, this multimedia file is divided into multiple multi-media segment, a multi-media segment corresponds to a story of a play or opera, according to the temporal information of the plurality of multi-media segment with this multimedia file, obtain the start time information of the plurality of multi-media segment.
The third providing in the disclosure may in implementation, this key message acquisition module 303 for:
For each multi-media segment, according to each frame picture of this multi-media segment and the lines information of this each frame picture, by face recognition technology, obtain the Role Information corresponding with the lines information of this each frame picture;
The Role Information corresponding according to the lines information with this each frame picture and the lines information of this each frame picture, generate the key message of this multi-media segment.
The 4th kind that provides in the disclosure may in implementation, this key message acquisition module 303 for:
Set up the character information database of this multimedia file, this character information database at least comprises Role Information in this multimedia file and face characteristic information corresponding to each role;
Each lines information corresponding according to this multi-media segment, obtains the source picture with this each lines information from this multi-media segment;
Obtain the face characteristic information in this source picture;
According to the face characteristic information in this source picture, from this character information database, obtain the Role Information that each lines information is corresponding with this.
In the 5th kind of possibility implementation that the disclosure provides, the key message of this multi-media segment also comprises the key picture of this multi-media segment.
In the 6th kind of possibility implementation that the disclosure provides, this device also comprises:
Labeling position acquisition module, for the start time information according to this multi-media segment, obtains the labeling position in playing progress bar;
Display module, for when the trigger action to labeling position in this playing progress bar being detected, shows the key message of corresponding multi-media segment.
About the device in above-described embodiment, wherein the concrete mode of modules executable operations has been described in detail in about the embodiment of the method, will not elaborate explanation herein.
Fig. 4 is the block diagram of a kind of information labeling device 400 according to an exemplary embodiment.Such as, device 400 may be provided in a server.With reference to Fig. 4, device 400 comprises processing components 422, and it comprises one or more processor further, and the memory resource representated by storer 432, can such as, by the instruction of the execution of processing element 422, application program for storing.The application program stored in storer 432 can comprise each module corresponding to one group of instruction one or more.In addition, processing components 422 is configured to perform instruction, to perform above-mentioned information labeling method.
Device 400 can also comprise the power management that a power supply module 426 is configured to actuating unit 400, and a wired or wireless network interface 450 is configured to device 400 to be connected to network, and input and output (I/O) interface 458.Device 400 can operate the operating system based on being stored in storer 432, such as WindowsServertM, MacOSXtM, UnixtM, LinuxtM, FreeBSDtMor it is similar.
Those skilled in the art, at consideration instructions and after putting into practice invention disclosed herein, will easily expect other embodiment of the present disclosure.The application is intended to contain any modification of the present disclosure, purposes or adaptations, and these modification, purposes or adaptations are followed general principle of the present disclosure and comprised the undocumented common practise in the art of the disclosure or conventional techniques means.Instructions and embodiment are only regarded as exemplary, and true scope of the present disclosure and spirit are pointed out by claim below.
Should be understood that, the disclosure is not limited to precision architecture described above and illustrated in the accompanying drawings, and can carry out various amendment and change not departing from its scope.The scope of the present disclosure is only limited by appended claim.