技术领域Technical Field
本申请涉及舞蹈教学技术领域，尤其涉及一种大学舞蹈考试测评方法、装置、设备及存储介质。The present application relates to the technical field of dance teaching, and in particular to a method, device, equipment and storage medium for evaluating a college dance examination.
背景技术Background technique
时代的不断发展, 互联网普遍被应用。传统的舞蹈教学模式也随之创新发展, 通过互联网信息技术将传统的舞蹈教学进行改革创新, 从而培养学生的的个性发展、自主学习和创新能力。让学生通过互联网进行自主学习，培养学生成为全面发展的优秀舞蹈人才。如今新媒体的发展, 为舞蹈教学的创新带来了途径与方式, 互联网的应用, 为舞蹈教学的改革与创新带来了机遇。科技与互联网的发展打破了传统的舞蹈教学, 促进了传统舞蹈教学的改革和创新。With the continuous development of the times, the Internet is widely used. The traditional dance teaching model has also been innovatively developed. Through the Internet information technology, the traditional dance teaching is reformed and innovated, so as to cultivate students' personality development, independent learning and innovation ability. Let students learn independently through the Internet and cultivate students to become well-rounded dance talents. Today, the development of new media has brought ways and methods for the innovation of dance teaching, and the application of the Internet has brought opportunities for the reform and innovation of dance teaching. The development of science and technology and the Internet has broken the traditional dance teaching and promoted the reform and innovation of traditional dance teaching.
目前的舞蹈测评方法，主要是传统的测评方式，由测试学生在考场进行舞蹈，舞蹈老师根据学生表现进行现场打分，这种方式较为费时，且测评结果容易带入测评老师的个人主观影响，不够公正。由此可知，现有技术在进行舞蹈测评时，存在耗费时间和测评老师容易带入个人主观因素的问题。The current dance assessment method is mainly the traditional assessment method, in which the test students dance in the examination room and the dance teacher scores the students on the spot according to their performance. This method is time-consuming and the assessment results are easily influenced by the personal subjective factors of the assessment teacher, which is not fair. It can be seen that the existing technology has the problems of time-consuming and personal subjective factors of the assessment teacher when conducting dance assessment.
发明内容Summary of the invention
本申请实施例的目的在于提出一种大学舞蹈考试测评方法、装置、设备及存储介质，以解决现有技术中在进行舞蹈测评时，存在耗费时间和测评老师容易带入个人主观因素的问题。The purpose of the embodiments of the present application is to propose a university dance examination assessment method, device, equipment and storage medium to solve the problems in the prior art that dance assessment is time-consuming and the assessment teacher is prone to bring in personal subjective factors.
为了解决上述技术问题，本申请实施例提供一种大学舞蹈考试测评方法，采用了如下所述的技术方案：In order to solve the above technical problems, the present application embodiment provides a college dance examination evaluation method, which adopts the following technical solution:
一种大学舞蹈考试测评方法，包括：A college dance examination and assessment method, comprising:
采集学生舞蹈考试的现场视频；Collect live video of students' dance exams;
获取网络上只标注动作类别的大量舞蹈动作图片，构建舞蹈图片集；Obtain a large number of dance action pictures with only action categories marked on the Internet and build a dance picture collection;
将所述舞蹈图片集作为训练集，传入预设的TLSVM模型，识别出所述训练集中不同动作种类；The dance picture set is used as a training set, and is passed into a preset TLSVM model to identify different types of movements in the training set;
获取所述不同动作种类对应的图像特征空间，并基于随机聚类森林的线性变换法，获取所述图像特征空间映射的视频特征空间，确定每一个所述视频特征空间对应的时空区域，构建所述训练集对应的时空区域集合；Obtain image feature spaces corresponding to the different action types, and based on the linear transformation method of random clustering forest, obtain video feature spaces mapped by the image feature spaces, determine the spatiotemporal regions corresponding to each of the video feature spaces, and construct a set of spatiotemporal regions corresponding to the training set;
将所述现场视频作为测试集，获取所述测试集中每一个动作对应的视频特征空间，确定所述视频特征空间对应的时空区域，作为测试时空区域；The live video is used as a test set, a video feature space corresponding to each action in the test set is obtained, and a spatiotemporal region corresponding to the video feature space is determined as a test spatiotemporal region;
基于所述时空区域集合，确定所述测试时空区域对应的时空区域，基于所述时空区域识别所述现场视频中不同的舞蹈动作；Based on the set of spatiotemporal regions, determining a spatiotemporal region corresponding to the test spatiotemporal region, and identifying different dance movements in the live video based on the spatiotemporal region;
将所述现场视频中识别出舞蹈动作与预设的测试舞蹈动作进行对比，并基于预设算法确定两者间的相似度，将所述相似度作为测评结果，完成大学舞蹈考试测评。The dance movements identified in the live video are compared with preset test dance movements, and the similarity between the two is determined based on a preset algorithm. The similarity is used as an evaluation result to complete the university dance test evaluation.
进一步的，所述采集学生舞蹈考试的现场视频，包括：Furthermore, the collection of live video of students' dance exams includes:
使用拍摄的方式在学生舞蹈考试时进行舞蹈视频录制，将录制内容作为采集的现场视频。Dance videos are recorded during students' dance exams by filming, and the recorded content is used as the collected live video.
进一步的，所述获取网络上只标注动作类别的大量舞蹈动作图片，包括：Furthermore, the method of obtaining a large number of dance action pictures on the Internet with only action categories marked includes:
将舞蹈动作图片作为关键词，基于大数据检索的方式在互联网上进行检索，下载大量所述舞蹈动作图片。The dance movement pictures are used as keywords, and a search is performed on the Internet based on a big data search method to download a large number of the dance movement pictures.
进一步的，所述预设的TLSVM模型，识别出所述训练集中不同动作种类，包括步骤如下：Furthermore, the preset TLSVM model identifies different action types in the training set, including the following steps:
对所述训练集中不同舞蹈图片，进行人物图像获取，确定所述人物图像中的不同肢体位置；For different dance pictures in the training set, character images are acquired to determine different limb positions in the character images;
基于所述不同肢体位置，构建肢体位置集，其中，所述肢体位置集包括：左手集、右手集、左腿集、右腿集、头部集、人体躯干集；Based on the different limb positions, construct a limb position set, wherein the limb position set includes: a left hand set, a right hand set, a left leg set, a right leg set, a head set, and a human body trunk set;
基于所述肢体位置集，确定所述不同肢体位置对应的时空区域；Based on the limb position set, determining the spatiotemporal regions corresponding to the different limb positions;
通过判定所述不同肢体位置对应的所述时空区域，识别出所述训练集中不同动作种类。By determining the space-time regions corresponding to the different limb positions, different action types in the training set are identified.
进一步的，所述获取所述不同动作种类对应的图像特征空间，包括：Furthermore, the obtaining of the image feature space corresponding to the different action types includes:
获取所述不同动作种类对应的不同肢体位置，基于Hessian矩阵的方式，识别所述不同肢体对应空间的极值；Obtaining different limb positions corresponding to the different action types, and identifying the extreme values of the space corresponding to the different limbs based on the Hessian matrix;
基于所述极值，构建不同肢体对应的尺度空间；Based on the extreme values, constructing scale spaces corresponding to different limbs;
获取所述尺度空间中的特征点，并进行过滤，对所述特征点进行精确定位；Acquire feature points in the scale space, filter them, and accurately locate the feature points;
获取所述不同特征点的主方向和所述不同特征点的特征值，基于所述主方向和特征值构建所述不同特征点对应形状特征，将所述形状特征作为所述不同动作种类对应的图像特征空间。The main directions of the different feature points and the feature values of the different feature points are obtained, shape features corresponding to the different feature points are constructed based on the main directions and the feature values, and the shape features are used as image feature spaces corresponding to the different action types.
进一步的，所述获取所述测试集中每一个动作对应的视频特征空间，包括：Furthermore, obtaining the video feature space corresponding to each action in the test set includes:
对所述测试集进行视频分割处理，将所述测试集分割成连贯的图像；Performing video segmentation processing on the test set to segment the test set into coherent images;
获取所述图像的颜色特征，纹理特征，人物形状特征；Acquire color features, texture features, and character shape features of the image;
分别获取所述图像的图像特征空间，并基于随机聚类森林的线性变换法，获取所述图像特征空间映射的视频特征空间。The image feature spaces of the images are respectively obtained, and based on the linear transformation method of random clustering forest, the video feature space mapped by the image feature space is obtained.
进一步的，所述确定所述视频特征空间对应的时空区域，作为测试时空区域，包括：Further, the determining the spatiotemporal region corresponding to the video feature space as the test spatiotemporal region includes:
获取视频中人物对应的二维图像块；Get the two-dimensional image blocks corresponding to the characters in the video;
基于所述视频时间顺序，对所述二维图像块进行三维图像块构建；Based on the video time sequence, constructing a three-dimensional image block from the two-dimensional image block;
将所述三维图像块对应的立方体块，作为所述视频特征空间对应的测试时空区域。The cubic block corresponding to the three-dimensional image block is used as the test space-time region corresponding to the video feature space.
进一步的，所述基于所述时空区域集合，确定所述测试时空区域对应的时空区域，包括：Further, the determining the space-time region corresponding to the test space-time region based on the space-time region set includes:
使用对比的方式，判断所述测试时空区域对应的时空区域。By using a comparison method, the space-time area corresponding to the test space-time area is determined.
进一步的，所述基于预设算法确定两者间的相似度，将所述相似度作为测评结果，完成大学舞蹈考试测评，包括：Furthermore, the similarity between the two is determined based on a preset algorithm, and the similarity is used as an evaluation result to complete the university dance test evaluation, including:
若测评学生为单人测评，则直接将所述现场视频中识别出舞蹈动作与预设的测试舞蹈动作进行对比，判断动作的相似度，若所述相似度超过预设的测评合格阈值，则所述学生测评结果合格，否则，所述学生测评结果不合格；If the student being evaluated is a single person, the dance movements identified in the live video are directly compared with the preset test dance movements to determine the similarity of the movements. If the similarity exceeds the preset evaluation pass threshold, the student's evaluation result is qualified; otherwise, the student's evaluation result is unqualified;
若测评学生为小组测评，则获取所述现场视频中识别出舞蹈动作与预设的测试舞蹈动作进行对比，判断小组中所有学生的动作相似度，并进行相似度平均值获取，若所述相似度平均值超过预设的测评合格阈值，则所述小组测评结果合格，否则，所述测评学生测评结果不合格。If the students being evaluated are in a group, the dance movements identified in the live video are obtained and compared with the preset test dance movements, the movement similarities of all students in the group are judged, and the average similarity value is obtained. If the average similarity value exceeds the preset evaluation pass threshold, the group evaluation result is qualified; otherwise, the student evaluation result is unqualified.
为了解决上述技术问题，本申请实施例还提供了一种大学舞蹈考试测评装置，采用了如下所述的技术方案：In order to solve the above technical problems, the embodiment of the present application also provides a college dance examination and evaluation device, which adopts the following technical solution:
一种大学舞蹈考试测评装置，包括：A university dance examination and evaluation device, comprising:
视频采集模块，用于采集学生舞蹈考试的现场视频；Video acquisition module, used to collect live videos of students' dance exams;
舞蹈图片集构建模块，用于获取网络上只标注动作类别的大量舞蹈动作图片，构建舞蹈图片集；The dance picture collection construction module is used to obtain a large number of dance action pictures with only action categories marked on the Internet to construct a dance picture collection;
动作种类识别模块，用于将所述舞蹈图片集作为训练集，传入预设的TLSVM模型，识别出所述训练集中不同动作种类；An action type recognition module is used to use the dance picture set as a training set, pass it into a preset TLSVM model, and recognize different action types in the training set;
训练时空区域集合生成模块，用于获取所述不同动作种类对应的图像特征空间，并基于随机聚类森林的线性变换法，获取所述图像特征空间映射的视频特征空间，确定每一个所述视频特征空间对应的时空区域，构建所述训练集对应的时空区域集合；A training spatiotemporal region set generation module is used to obtain the image feature space corresponding to the different action types, and based on the linear transformation method of random clustering forest, obtain the video feature space mapped by the image feature space, determine the spatiotemporal region corresponding to each of the video feature spaces, and construct the spatiotemporal region set corresponding to the training set;
测试时空区域确定模块，用于将所述现场视频作为测试集，获取所述测试集中每一个动作对应的视频特征空间，确定所述视频特征空间对应的时空区域，作为测试时空区域；A test spatiotemporal region determination module, used to use the live video as a test set, obtain a video feature space corresponding to each action in the test set, and determine a spatiotemporal region corresponding to the video feature space as a test spatiotemporal region;
舞蹈动作识别模块，用于基于所述时空区域集合，确定所述测试时空区域对应的时空区域，基于所述时空区域识别所述现场视频中不同的舞蹈动作；A dance movement recognition module, used to determine the space-time region corresponding to the test space-time region based on the space-time region set, and recognize different dance movements in the live video based on the space-time region;
舞蹈考试测评模块，用于将所述现场视频中识别出舞蹈动作与预设的测试舞蹈动作进行对比，并基于预设算法确定两者间的相似度，将所述相似度作为测评结果，完成大学舞蹈考试测评。The dance test evaluation module is used to compare the dance movements identified in the live video with the preset test dance movements, and determine the similarity between the two based on a preset algorithm, and use the similarity as the evaluation result to complete the university dance test evaluation.
为了解决上述技术问题，本申请实施例还提供一种计算机设备，采用了如下所述的技术方案：In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:
一种计算机设备，包括存储器和处理器，所述存储器中存储有计算机程序，所述处理器执行所述计算机程序时实现本申请实施例中提出的一种大学舞蹈考试测评方法的步骤。A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of a college dance examination evaluation method proposed in an embodiment of the present application are implemented.
为了解决上述技术问题，本申请实施例还提供一种非易失性计算机可读存储介质，采用了如下所述的技术方案：In order to solve the above technical problems, the embodiment of the present application further provides a non-volatile computer-readable storage medium, which adopts the following technical solution:
一种非易失性计算机可读存储介质，计算机可读存储介质上存储有计算机程序，所述计算机程序被处理器执行时实现本申请实施例中提出的一种大学舞蹈考试测评方法的步骤。A non-volatile computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of a college dance examination evaluation method proposed in an embodiment of the present application are implemented.
与现有技术相比，本申请实施例主要有以下有益效果：Compared with the prior art, the embodiments of the present application have the following beneficial effects:
本申请实施例公开了大学舞蹈考试测评方法、装置、设备及存储介质，通过采集学生舞蹈考试的现场视频；构建舞蹈图片集；将舞蹈图片集传入预设的TLSVM模型，识别出不同动作种类；获取不同动作种类对应的时空区域，构建所述训练集对应的时空区域集合；获取现场视频中每一个动作对应的测试时空区域；从训练时空区域集合中确定测试时空区域一一对应的时空区域，基于时空区域识别现场视频中的舞蹈动作；将现场视频中识别出舞蹈动作与预设的测试舞蹈动作进行对比，确定两者间的相似度，将相似度作为测评结果，完成大学舞蹈考试测评。本申请避免监考老师的主观因素影响，做到公平进行舞蹈测评。The embodiment of the present application discloses a method, device, equipment and storage medium for evaluating a university dance exam, which collects live videos of students' dance exams; constructs a dance picture set; passes the dance picture set into a preset TLSVM model to identify different types of movements; obtains the spatiotemporal regions corresponding to different types of movements, and constructs a spatiotemporal region set corresponding to the training set; obtains the test spatiotemporal region corresponding to each movement in the live video; determines the spatiotemporal region corresponding to the test spatiotemporal region from the training spatiotemporal region set, and identifies the dance movements in the live video based on the spatiotemporal region; compares the dance movements identified in the live video with the preset test dance movements, determines the similarity between the two, and uses the similarity as the evaluation result to complete the evaluation of the university dance exam. The present application avoids the influence of subjective factors of the invigilator and ensures fair dance evaluation.
附图说明BRIEF DESCRIPTION OF THE DRAWINGS
为了更清楚地说明本申请中的方案，下面将对本申请实施例描述中所需要使用的附图作一个简单介绍，显而易见地，下面描述中的附图是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
图1为本申请实施例可以应用于其中的示例性系统架构图；FIG1 is a diagram of an exemplary system architecture to which embodiments of the present application may be applied;
图2为本申请实施例中所述大学舞蹈考试测评方法的一个实施例的流程图；FIG2 is a flow chart of an embodiment of the university dance examination evaluation method described in the embodiments of the present application;
图3为本申请实施例中所述大学舞蹈考试测评方法的一个实施例的执行逻辑图；FIG3 is an execution logic diagram of an embodiment of the university dance examination evaluation method described in the embodiments of the present application;
图4为本申请实施例中所述大学舞蹈考试测评装置的一个实施例的结构示意图；FIG4 is a schematic structural diagram of an embodiment of a university dance examination and evaluation device according to an embodiment of the present application;
图5为本申请实施例中计算机设备的一个实施例的结构示意图。FIG. 5 is a schematic diagram of the structure of a computer device according to an embodiment of the present application.
具体实施方式Detailed ways
除非另有定义，本文所使用的所有的技术和科学术语与属于本申请的技术领域的技术人员通常理解的含义相同；本文中在申请的说明书中所使用的术语只是为了描述具体的实施例的目的，不是旨在于限制本申请；本申请的说明书和权利要求书及上述附图说明中的术语“包括”和“具有”以及它们的任何变形，意图在于覆盖不排他的包含。本申请的说明书和权利要求书或上述附图中的术语“第一”、“第二”等是用于区别不同对象，而不是用于描述特定顺序。Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.
在本文中提及“实施例”意味着，结合实施例描述的特定特征、结构或特性可以包含在本申请的至少一个实施例中。在说明书中的各个位置出现该短语并不一定均是指相同的实施例，也不是与其它实施例互斥的独立的或备选的实施例。本领域技术人员显式地和隐式地理解的是，本文所描述的实施例可以与其它实施例相结合。Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.
为了使本技术领域的人员更好地理解本申请方案，下面将结合附图，对本申请实施例中的技术方案进行清楚、完整地描述。In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.
如图1所示，系统架构100可以包括终端设备101、102、103，网络104和服务器105。网络104用以在终端设备101、102、103和服务器105之间提供通信链路的介质。网络104可以包括各种连接类型，例如有线、无线通信链路或者光纤电缆等等。As shown in Fig. 1, system architecture 100 may include terminal devices 101, 102, 103, network 104 and server 105. Network 104 is used to provide a medium for communication links between terminal devices 101, 102, 103 and server 105. Network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.
用户可以使用终端设备101、102、103通过网络104与服务器105交互，以接收或发送消息等。终端设备101、102、103上可以安装有各种通讯客户端应用，例如网页浏览器应用、购物类应用、搜索类应用、即时通信工具、邮箱客户端、社交平台软件等。Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.
终端设备101、102、103可以是具有显示屏并且支持网页浏览的各种电子设备，包括但不限于智能手机、平板电脑、电子书阅读器、MP3播放器(Moving Picture ExpertsGroup Audio Layer III，动态影像专家压缩标准音频层面3)、MP4(Moving PictureExperts Group Audio Layer IV，动态影像专家压缩标准音频层面4)播放器、膝上型便携计算机和台式计算机等等。Terminal devices 101, 102, 103 can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 (Moving Picture Experts Group Audio Layer IV), laptop computers, desktop computers, etc.
服务器105可以是提供各种服务的服务器，例如对终端设备101、102、103上显示的页面提供支持的后台服务器。The server 105 may be a server that provides various services, such as a background server that provides support for web pages displayed on the terminal devices 101 , 102 , and 103 .
需要说明的是，本申请实施例所提供的大学舞蹈考试测评方法一般由服务器/终端设备执行，相应地，大学舞蹈考试测评装置一般设置于服务器/终端设备中。It should be noted that the university dance examination evaluation method provided in the embodiment of the present application is generally executed by a server/terminal device, and accordingly, the university dance examination evaluation device is generally arranged in the server/terminal device.
应该理解，图1中的终端设备、网络和服务器的数目仅仅是示意性的。根据实现需要，可以具有任意数目的终端设备、网络和服务器。It should be understood that the number of terminal devices, networks and servers in Figure 1 is only illustrative. Any number of terminal devices, networks and servers may be provided according to implementation requirements.
继续参考图2，图中示出了本申请的大学舞蹈考试测评方法的一个实施例的流程图，所述的大学舞蹈考试测评方法包括以下步骤：Continuing to refer to FIG. 2 , a flow chart of an embodiment of a university dance examination and evaluation method of the present application is shown, and the university dance examination and evaluation method comprises the following steps:
步骤201，采集学生舞蹈考试的现场视频。Step 201, collecting live video of students' dance exams.
在本申请实施例中，所述采集学生舞蹈考试的现场视频，包括：使用拍摄的方式在学生舞蹈考试时进行舞蹈视频录制，将录制内容作为采集的现场视频。In an embodiment of the present application, the collecting of live video of the student dance test includes: recording the dance video during the student dance test by shooting, and using the recorded content as the collected live video.
步骤202，获取网络上只标注动作类别的大量舞蹈动作图片，构建舞蹈图片集。Step 202, obtaining a large number of dance action pictures with only action categories marked on the Internet, and constructing a dance picture collection.
在本申请实施例中，所述获取网络上只标注动作类别的大量舞蹈动作图片，包括：将舞蹈动作图片作为关键词，基于大数据检索的方式在互联网上进行检索，下载大量所述舞蹈动作图片。In an embodiment of the present application, obtaining a large number of dance action pictures on the Internet with only action categories marked includes: using the dance action pictures as keywords, searching on the Internet based on big data retrieval, and downloading a large number of the dance action pictures.
步骤203，将所述舞蹈图片集作为训练集，传入预设的TLSVM模型，识别出所述训练集中不同动作种类。Step 203: Use the dance picture set as a training set, input it into a preset TLSVM model, and identify different types of movements in the training set.
在本申请实施例中，所述预设的TLSVM模型，识别出所述训练集中不同动作种类，包括步骤如下：对所述训练集中不同舞蹈图片，进行人物图像获取，确定所述人物图像中的不同肢体位置；基于所述不同肢体位置，构建肢体位置集，其中，所述肢体位置集包括：左手集、右手集、左腿集、右腿集、头部集、人体躯干集；基于所述肢体位置集，确定所述不同肢体位置对应的时空区域；通过判定所述不同肢体位置对应的所述时空区域，识别出所述训练集中不同动作种类。In an embodiment of the present application, the preset TLSVM model identifies different types of actions in the training set, including the following steps: acquiring character images for different dance pictures in the training set, and determining different limb positions in the character images; constructing a limb position set based on the different limb positions, wherein the limb position set includes: a left hand set, a right hand set, a left leg set, a right leg set, a head set, and a human torso set; based on the limb position set, determining the space-time regions corresponding to the different limb positions; identifying different types of actions in the training set by determining the space-time regions corresponding to the different limb positions.
解释：所述对所述训练集中不同舞蹈图片，进行人物图像获取，确定所述人物图像中的不同肢体位置，具体实现方式如下：将舞蹈图片集中元素进行按比例压缩；使用GrabCut算法对照片进行处理，获取人物的掩像；对所述人物的掩像进行边缘检测，划分出绝对背景区域，并对绝对背景区域进行采样取色，计算其绝对背景的RGB值；对有背景残留的掩像分别单独进行腐蚀和膨胀的形态学处理，然后将两种形态学处理结果进行减法操作获取边缘区域，该边缘区域为边缘未知区域；将边缘未知区域的像素重新划分至背景或前景，划分方法是将绝对背景的RGB值与未知像素的颜色值进行比较，距离越近则属于背景的可能性越高，根据判断条件重新分割；对边缘平滑处理，最后输出的分割图像，即舞蹈图片中的人物图像；对所述人物图像进行单独分割，确定人物图像中不同肢体位置。Explanation: The method of acquiring character images from different dance pictures in the training set and determining different limb positions in the character images is specifically implemented as follows: compressing the elements in the dance picture set proportionally; processing the photos using the GrabCut algorithm to obtain the character's mask image; performing edge detection on the character's mask image to divide the absolute background area, sampling and coloring the absolute background area, and calculating the RGB value of the absolute background; performing morphological processing of corrosion and expansion on the mask image with background residues, and then subtracting the two morphological processing results to obtain the edge area, which is the edge unknown area; re-dividing the pixels in the edge unknown area into the background or foreground, and the division method is to compare the RGB value of the absolute background with the color value of the unknown pixel, the closer the distance, the higher the possibility of belonging to the background, and re-segmenting according to the judgment condition; performing edge smoothing, and finally outputting the segmented image, that is, the character image in the dance picture; segmenting the character image separately to determine different limb positions in the character image.
解释：所述基于所述不同肢体位置，构建肢体位置集，其中，所述肢体位置集包括：左手集、右手集、左腿集、右腿集、头部集、人体躯干集，具体方式如下：在对所述训练集中不同舞蹈图片，进行人物图像获取之前，对所述舞蹈图片集中元素进行区别标识，提取人物图像之后，将所述区别标识添加到人物图像中，再添加到所述不同肢体位置，将不同标识的同类肢体添加到一个集合中，构建肢体位置集。Explanation: Based on the different limb positions, a limb position set is constructed, wherein the limb position set includes: left hand set, right hand set, left leg set, right leg set, head set, and human torso set. The specific method is as follows: before obtaining character images from different dance pictures in the training set, the elements in the dance picture set are distinguished and identified. After extracting the character image, the distinguishing mark is added to the character image and then added to the different limb positions. The same type of limbs with different marks are added to a set to construct a limb position set.
例如：所述舞蹈图片集中存在1万个舞蹈图片，为了区分使用1到10000的阿拉伯数字作为图片名称；在提取到舞蹈图片中的人物图像后，也使用1到10000作为人物图像的图片名称，若舞蹈图片300中有三个人物，则使用300a，300b，300c进行区别标识；对舞蹈图片300进行肢体位置确定后，使用左手_300a、左手_300b、左手_300c、右手_300a、右手_300b、右手_300c、左腿_300a、左腿_300b、左腿_300c、右腿_300a、右腿_300b、右腿_300c、头部_300a、头部_300b、头部_300c、躯干_300a、躯干_300b、躯干_300c，将上述左手_300a、左手_300b、左手_300c与其他舞蹈图片人物图像中的左手图片，放入同一个集合，构建左手集，同理，构建右手集、左腿集、右腿集、头部集、人体躯干集。For example, there are 10,000 dance pictures in the dance picture collection. In order to distinguish them, Arabic numerals from 1 to 10,000 are used as picture names. After extracting the character images in the dance pictures, 1 to 10,000 are also used as picture names of the character images. If there are three characters in the dance picture 300, 300a, 300b, and 300c are used for distinguishing marks. After determining the limb positions of the dance picture 300, left hand_300a, left hand_300b, left hand_300c, right hand_300a, right hand_300b, and right hand_300c are used for distinguishing marks. _300c, left leg_300a, left leg_300b, left leg_300c, right leg_300a, right leg_300b, right leg_300c, head_300a, head_300b, head_300c, torso_300a, torso_300b, torso_300c, put the above left hand_300a, left hand_300b, left hand_300c and the left hand pictures in other dance pictures and characters into the same set to construct the left hand set. Similarly, construct the right hand set, left leg set, right leg set, head set, and human torso set.
解释：所述基于所述肢体位置集，确定所述不同肢体位置对应的时空区域，具体实现方式如下：确定所述不同肢体位置中边缘点，并获取边缘点的二维坐标，确定所述不同肢体位置的最小外包矩形，获取6个构成肢体长方体的最小外包矩形，生成所述肢体位置最小外包立方体，将所述最小外包立方体作为其对应的时空区域。Explanation: Based on the limb position set, the space-time areas corresponding to the different limb positions are determined. The specific implementation method is as follows: determine the edge points in the different limb positions, and obtain the two-dimensional coordinates of the edge points, determine the minimum bounding rectangles of the different limb positions, obtain the six minimum bounding rectangles that constitute the limb cuboid, generate the minimum bounding cube of the limb position, and use the minimum bounding cube as its corresponding space-time area.
解释：所述通过判定所述不同肢体位置对应的所述时空区域，识别出所述训练集中不同动作种类，具体实现方式如下：将人体躯干对应的时空区域作为基准区域，判断其他人体肢体位置对应的时空区域相对应基准区域的位置，从而识别出所述训练集中不同动作所属的种类。Explanation: The different types of actions in the training set are identified by determining the space-time regions corresponding to the different limb positions. The specific implementation method is as follows: the space-time region corresponding to the human torso is used as the reference region, and the positions of the space-time regions corresponding to other human limb positions relative to the reference region are determined, thereby identifying the types of different actions in the training set.
例如：某个舞蹈图片集中元素对应的人物图像，编号为31，再上述步骤中，已经分别获取到左手_31、右手_31、左腿_31、右腿_31、头部_31、躯干_31对应的时空区域，此时，将躯干_31对应的时空区域作为基准区域，若某个时空区域在其右上方，则为左手_31对应的时空区域，确定所述动作为左手种类，同理，可以识别出其他时空区域，确定不同动作所属的种类。For example: the character image corresponding to the element in a dance picture set is numbered 31. In the above steps, the space-time areas corresponding to the left hand_31, right hand_31, left leg_31, right leg_31, head_31, and torso_31 have been obtained respectively. At this time, the space-time area corresponding to the torso_31 is used as the reference area. If a space-time area is to the upper right of it, it is the space-time area corresponding to the left hand_31, and the action is determined to be of the left hand type. Similarly, other space-time areas can be identified to determine the types to which different actions belong.
步骤204，获取所述不同动作种类对应的图像特征空间，并基于随机聚类森林的线性变换法，获取所述图像特征空间映射的视频特征空间，确定每一个所述视频特征空间对应的时空区域，构建所述训练集对应的时空区域集合。Step 204, obtain the image feature space corresponding to the different action types, and based on the linear transformation method of random clustering forest, obtain the video feature space mapped by the image feature space, determine the space-time region corresponding to each of the video feature spaces, and construct a space-time region set corresponding to the training set.
在本申请实施例中，所述获取所述不同动作种类对应的图像特征空间，包括：获取所述不同动作种类对应的不同肢体位置，基于Hessian矩阵的方式，识别所述不同肢体对应空间的极值；基于所述极值，构建不同肢体对应的尺度空间；获取所述尺度空间中的特征点，并进行过滤，对所述特征点进行精确定位；获取所述不同特征点的主方向和所述不同特征点的特征值，基于所述主方向和特征值构建所述不同特征点对应形状特征，将所述形状特征作为所述不同动作种类对应的图像特征空间。In an embodiment of the present application, the obtaining of the image feature space corresponding to the different action types includes: obtaining different limb positions corresponding to the different action types, and identifying the extreme values of the space corresponding to the different limbs based on the Hessian matrix; constructing the scale space corresponding to the different limbs based on the extreme values; obtaining the feature points in the scale space, filtering them, and accurately locating the feature points; obtaining the main directions of the different feature points and the feature values of the different feature points, constructing the shape features corresponding to the different feature points based on the main directions and the feature values, and using the shape features as the image feature space corresponding to the different action types.
解释：所述获取所述不同动作种类对应的不同肢体位置，基于Hessian矩阵的方式，识别所述不同肢体对应空间的极值，具体方式如下：确定所述不同肢体在二维图像中的位置，并对不同肢体包含的每个像素点计算图像在X方向和Y方向的二阶偏导数，计算图像的XY方向的导数，所述肢体在临界点C(x,y)点的Hession矩阵为H(C),若所述H(C)为正定矩阵，则临界点C处是一个局部极小值，若所述H(C)为负定矩阵，则临界点C处是一个局部极大值，若所述H(C)为不定矩阵，则临界点C处不是极值，识别出的所有极大值和极小值对应的点构成所述不同肢体对应空间的极值。Explanation: The method of obtaining different limb positions corresponding to the different action types and identifying the extreme values of the space corresponding to the different limbs based on the Hessian matrix is as follows: determining the positions of the different limbs in the two-dimensional image, and calculating the second-order partial derivatives of the image in the X and Y directions for each pixel point contained in the different limbs, and calculating the derivative of the image in the XY direction. The Hession matrix of the limb at the critical point C(x,y) is H(C). If the H(C) is a positive definite matrix, the critical point C is a local minimum. If the H(C) is a negative definite matrix, the critical point C is a local maximum. If the H(C) is an indefinite matrix, the critical point C is not an extreme value. The points corresponding to all the identified maxima and minima constitute the extreme values of the space corresponding to the different limbs.
解释：所述基于所述极值，构建不同肢体对应的尺度空间，具体方式如下：上述步骤获取到若干个极值点，通过这些极值点集，确定不同肢体对应的尺度空间。Explanation: Based on the extreme values, the scale space corresponding to different limbs is constructed, and the specific method is as follows: the above steps obtain a number of extreme value points, and the scale space corresponding to different limbs is determined through these extreme value point sets.
解释：所述获取所述尺度空间中的特征点，并进行过滤，对所述特征点进行精确定位，具体实现方式如下：将所述极值点集中的点作为所述尺度空间中的特征点，并将所述极大值的点作为极大特征值点，将所述极小值的点作为极小特征值点，进行区分，并分别获取极大特征值点对应的点集和极小特征值点对应的点集。Explanation: The feature points in the scale space are obtained and filtered to accurately locate the feature points. The specific implementation method is as follows: the points in the extreme value point set are used as feature points in the scale space, and the points of the maximum value are used as maximum eigenvalue points, and the points of the minimum value are used as minimum eigenvalue points, and they are distinguished, and the point set corresponding to the maximum eigenvalue points and the point set corresponding to the minimum eigenvalue points are respectively obtained.
解释：所述获取所述不同特征点的主方向和所述不同特征点的特征值，基于所述主方向和特征值构建所述不同特征点对应形状特征，将所述形状特征作为所述不同动作种类对应的图像特征空间，具体的实现方式如下：分别将相邻的极大值点和相邻的极小值点进行点间连接，构成点线，若由极大值点构成的点线两侧分别存在一条由极小值点组成的点线，则极大值点构成的点线方向即为所述特征点的主方向，获取所有极大值点进行构建的若干条点线，即为图像的形状线条，将所述形状线条组成的二维空间区域作为所述不同动作种类对应的图像特征空间。Explanation: The main directions of the different feature points and the feature values of the different feature points are obtained, shape features corresponding to the different feature points are constructed based on the main directions and the feature values, and the shape features are used as the image feature space corresponding to the different action types. The specific implementation method is as follows: adjacent maximum points and adjacent minimum points are connected point by point to form point lines. If there is a point line composed of minimum points on both sides of the point line composed of maximum points, then the direction of the point line composed of maximum points is the main direction of the feature points. A number of point lines constructed by all maximum points are obtained, which are the shape lines of the image. The two-dimensional space area composed of the shape lines is used as the image feature space corresponding to the different action types.
解释：所述并基于随机聚类森林的线性变换法，获取所述图像特征空间映射的视频特征空间，具体实现方式如下：从上述图像的形状线条中采取有放回的抽样，构造子线条集，子线条集的线条数量与上述图像的形状线条数量相同；不同子线条集中的线条可以重复，同一个子线条集中的元素也可以重复；利用子线条集来构建子决策树，将所述图像特征空间对应的所有线条构成空间线条集，并将所述空间线条集放到每个子决策树中，每个子决策树输出一个空间构建结果；若大部分子决策树的输出结果一致，则将占比例最大的空间构建结果，作为随机聚类森林结果。若所述随机聚类森林结果，在经过矩阵变换之后，改变的是组成向量的基，而所述随机聚类森林结果关于基的线性组合方式没有变化，则将所述随机聚类森林结果作为所述图像特征空间映射的视频特征空间。Explanation: The linear transformation method based on random clustering forest is used to obtain the video feature space mapped by the image feature space. The specific implementation method is as follows: sampling with replacement is taken from the shape lines of the above image to construct a sub-line set, and the number of lines in the sub-line set is the same as the number of shape lines of the above image; lines in different sub-line sets can be repeated, and elements in the same sub-line set can also be repeated; sub-decision trees are constructed using sub-line sets, and all lines corresponding to the image feature space constitute a spatial line set, and the spatial line set is placed in each sub-decision tree, and each sub-decision tree outputs a spatial construction result; if the output results of most sub-decision trees are consistent, the spatial construction result with the largest proportion is used as the random clustering forest result. If the random clustering forest result, after the matrix transformation, changes the basis of the constituent vectors, and the linear combination method of the random clustering forest result with respect to the basis does not change, then the random clustering forest result is used as the video feature space mapped by the image feature space.
步骤205，将所述现场视频作为测试集，获取所述测试集中每一个动作对应的视频特征空间，确定所述视频特征空间对应的时空区域，作为测试时空区域。Step 205, taking the live video as a test set, obtaining the video feature space corresponding to each action in the test set, and determining the spatiotemporal region corresponding to the video feature space as the test spatiotemporal region.
在本申请实施例中，所述获取所述测试集中每一个动作对应的视频特征空间，包括：对所述测试集进行视频分割处理，将所述测试集分割成连贯的图像；获取所述图像的颜色特征，纹理特征，人物形状特征；分别获取所述图像的图像特征空间，并基于随机聚类森林的线性变换法，获取所述图像特征空间映射的视频特征空间。In an embodiment of the present application, obtaining the video feature space corresponding to each action in the test set includes: performing video segmentation processing on the test set to segment the test set into coherent images; obtaining color features, texture features, and character shape features of the image; respectively obtaining the image feature space of the image, and obtaining the video feature space mapped by the image feature space based on the linear transformation method of random clustering forest.
在本申请实施例中，所述确定所述视频特征空间对应的时空区域，作为测试时空区域，包括：获取视频中人物对应的二维图像块；基于所述视频时间顺序，对所述二维图像块进行三维图像块构建；将所述三维图像块对应的立方体块，作为所述视频特征空间对应的测试时空区域。In an embodiment of the present application, determining the spatiotemporal region corresponding to the video feature space as a test spatiotemporal region includes: obtaining two-dimensional image blocks corresponding to characters in the video; constructing three-dimensional image blocks from the two-dimensional image blocks based on the video time sequence; and using the cubic blocks corresponding to the three-dimensional image blocks as the test spatiotemporal region corresponding to the video feature space.
解释：在上述步骤中已经将所述测试集进行视频分割处理，将所述测试集分割成连贯的图像，此时，分别获取不同图像中人物占据的二维空间，并通过视频时间顺序，对所述二维空间进行叠加，进行三维图像块构建，所述三维图像对应的立方体快，即为舞蹈视频中人物进行舞蹈时占据的空间立方体，作为所述视频特征空间对应的时空子区域。Explanation: In the above steps, the test set has been subjected to video segmentation processing and divided into coherent images. At this time, the two-dimensional spaces occupied by the characters in different images are obtained respectively, and the two-dimensional spaces are superimposed through the video time sequence to construct three-dimensional image blocks. The cube block corresponding to the three-dimensional image is the space cube occupied by the characters in the dance video when dancing, which serves as the space-time sub-region corresponding to the video feature space.
步骤206，基于所述时空区域集合，确定所述测试时空区域对应的时空区域，基于所述时空区域识别所述现场视频中不同的舞蹈动作。Step 206: Based on the set of spatiotemporal regions, determine the spatiotemporal region corresponding to the test spatiotemporal region, and identify different dance movements in the live video based on the spatiotemporal region.
在本申请实施例中，所述基于所述时空区域集合，确定所述测试时空区域对应的时空区域，包括：使用对比的方式，判断所述测试时空区域对应的时空区域。In the embodiment of the present application, determining the space-time region corresponding to the test space-time region based on the space-time region set includes: using a comparison method to determine the space-time region corresponding to the test space-time region.
步骤207，将所述现场视频中识别出舞蹈动作与预设的测试舞蹈动作进行对比，并基于预设算法确定两者间的相似度，将所述相似度作为测评结果，完成大学舞蹈考试测评。Step 207, comparing the dance movements identified in the live video with the preset test dance movements, and determining the similarity between the two based on a preset algorithm, taking the similarity as the evaluation result, and completing the college dance test evaluation.
在本申请实施例中，所述基于预设算法确定两者间的相似度，将所述相似度作为测评结果，完成大学舞蹈考试测评，包括：若测评学生为单人测评，则直接将所述现场视频中识别出舞蹈动作与预设的测试舞蹈动作进行对比，判断动作的相似度，若所述相似度超过预设的测评合格阈值，则所述学生测评结果合格，否则，所述学生测评结果不合格；若测评学生为小组测评，则获取所述现场视频中识别出舞蹈动作与预设的测试舞蹈动作进行对比，判断小组中所有学生的动作相似度，并进行相似度平均值获取，若所述相似度平均值超过预设的测评合格阈值，则所述小组测评结果合格，否则，所述测评学生测评结果不合格。In an embodiment of the present application, the similarity between the two is determined based on a preset algorithm, and the similarity is used as an evaluation result to complete the university dance test evaluation, including: if the student being evaluated is a single person, the dance movements identified in the live video are directly compared with the preset test dance movements to determine the similarity of the movements. If the similarity exceeds the preset evaluation pass threshold, the student evaluation result is qualified, otherwise, the student evaluation result is unqualified; if the student being evaluated is a group evaluation, the dance movements identified in the live video are obtained and compared with the preset test dance movements to determine the movement similarity of all students in the group, and the similarity average value is obtained. If the similarity average value exceeds the preset evaluation pass threshold, the group evaluation result is qualified, otherwise, the student evaluation result is unqualified.
继续参考图3，图3为本申请实施例中所述大学舞蹈考试测评方法的一个实施例的执行逻辑图，具体执行步骤如下：采集学生舞蹈考试的现场视频；获取网络上只标注动作类别的大量舞蹈动作图片，构建舞蹈图片集；将所述舞蹈图片集作为训练集，传入预设的TLSVM模型，识别出所述训练集中不同动作种类；获取所述不同动作种类对应的图像特征空间，并基于随机聚类森林的线性变换法，获取所述图像特征空间映射的视频特征空间，确定每一个所述视频特征空间对应的时空区域，构建所述训练集对应的时空区域集合；将所述现场视频作为测试集，获取所述测试集中每一个动作对应的视频特征空间，确定所述视频特征空间对应的时空区域，作为测试时空区域；基于所述时空区域集合，确定所述测试时空区域对应的时空区域，基于所述时空区域识别所述现场视频中不同的舞蹈动作；将所述现场视频中识别出舞蹈动作与预设的测试舞蹈动作进行对比，并基于预设算法确定两者间的相似度，将所述相似度作为测评结果，完成大学舞蹈考试测评。Continuing to refer to Figure 3, Figure 3 is an execution logic diagram of an embodiment of the university dance test evaluation method described in the embodiment of the present application, and the specific execution steps are as follows: collecting live videos of student dance tests; obtaining a large number of dance action pictures with only action categories marked on the Internet to construct a dance picture set; using the dance picture set as a training set, passing it into a preset TLSVM model, and identifying different types of actions in the training set; obtaining the image feature space corresponding to the different types of actions, and based on the linear transformation method of random clustering forests, obtaining the video feature space mapped by the image feature space, determining the spatiotemporal region corresponding to each of the video feature spaces, and constructing a spatiotemporal region set corresponding to the training set; using the live video as a test set, obtaining the video feature space corresponding to each action in the test set, and determining the spatiotemporal region corresponding to the video feature space as a test spatiotemporal region; based on the spatiotemporal region set, determining the spatiotemporal region corresponding to the test spatiotemporal region, and identifying different dance actions in the live video based on the spatiotemporal region; comparing the dance actions identified in the live video with the preset test dance actions, and determining the similarity between the two based on a preset algorithm, using the similarity as an evaluation result, and completing the university dance test evaluation.
本申请实施例中所述的大学舞蹈考试测评方法，可以通过采集学生舞蹈考试的现场视频；构建舞蹈图片集；将舞蹈图片集传入预设的TLSVM模型，识别出不同动作种类；获取不同动作种类对应的时空区域，构建所述训练集对应的时空区域集合；获取现场视频中每一个动作对应的测试时空区域；从训练时空区域集合中确定测试时空区域一一对应的时空区域，基于时空区域识别现场视频中的舞蹈动作；将现场视频中识别出舞蹈动作与预设的测试舞蹈动作进行对比，确定两者间的相似度，将相似度作为测评结果，完成大学舞蹈考试测评。本申请避免监考老师的主观因素影响，做到公平进行舞蹈测评。The university dance test evaluation method described in the embodiment of the present application can collect live videos of students' dance tests; construct a dance picture set; pass the dance picture set into a preset TLSVM model to identify different types of movements; obtain the spatiotemporal regions corresponding to different types of movements, and construct a set of spatiotemporal regions corresponding to the training set; obtain the test spatiotemporal regions corresponding to each movement in the live video; determine the spatiotemporal regions corresponding to the test spatiotemporal regions from the training spatiotemporal region set, and identify the dance movements in the live video based on the spatiotemporal regions; compare the dance movements identified in the live video with the preset test dance movements, determine the similarity between the two, and use the similarity as the evaluation result to complete the university dance test evaluation. This application avoids the influence of subjective factors of the invigilator and ensures fair dance evaluation.
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，该计算机程序可存储于一计算机可读取存储介质中，该程序在执行时，可包括如上述各方法的实施例的流程。其中，前述的存储介质可为磁碟、光盘、只读存储记忆体（Read-Only Memory，ROM）等非易失性存储介质，或随机存储记忆体（Random Access Memory，RAM）等。Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).
应该理解的是，虽然附图的流程图中的各个步骤按照箭头的指示依次显示，但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明，这些步骤的执行并没有严格的顺序限制，其可以以其他的顺序执行。而且，附图的流程图中的至少一部分步骤可以包括多个子步骤或者多个阶段，这些子步骤或者阶段并不必然是在同一时刻执行完成，而是可以在不同的时刻执行，其执行顺序也不必然是依次进行，而是可以与其他步骤或者其他步骤的子步骤或者阶段的至少一部分轮流或者交替地执行。It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.
进一步参考图4，作为对上述图2所示方法的实现，本申请提供了一种大学舞蹈考试测评装置的一个实施例，该装置实施例与图2所示的方法实施例相对应，该装置具体可以应用于各种电子设备中。Further referring to FIG. 4 , as an implementation of the method shown in FIG. 2 , the present application provides an embodiment of a university dance examination and evaluation device. The device embodiment corresponds to the method embodiment shown in FIG. 2 , and the device can be specifically applied to various electronic devices.
如图4所示，本实施例所述的大学舞蹈考试测评装置4包括：视频采集模块401、舞蹈图片集构建模块402、动作种类识别模块403、训练时空区域集合生成模块404、测试时空区域确定模块405、舞蹈动作识别模块406、舞蹈考试测评模块407。其中：As shown in FIG4 , the university dance test evaluation device 4 described in this embodiment includes: a video acquisition module 401, a dance picture set construction module 402, an action type recognition module 403, a training space-time region set generation module 404, a test space-time region determination module 405, a dance action recognition module 406, and a dance test evaluation module 407. Among them:
视频采集模块401，用于采集学生舞蹈考试的现场视频；The video acquisition module 401 is used to acquire the on-site video of the student dance test;
舞蹈图片集构建模块402，用于获取网络上只标注动作类别的大量舞蹈动作图片，构建舞蹈图片集；The dance picture collection construction module 402 is used to obtain a large number of dance action pictures with only action categories marked on the Internet to construct a dance picture collection;
动作种类识别模块403，用于将所述舞蹈图片集作为训练集，传入预设的TLSVM模型，识别出所述训练集中不同动作种类；The action type identification module 403 is used to use the dance picture set as a training set, input it into a preset TLSVM model, and identify different action types in the training set;
训练时空区域集合生成模块404，用于获取所述不同动作种类对应的图像特征空间，并基于随机聚类森林的线性变换法，获取所述图像特征空间映射的视频特征空间，确定每一个所述视频特征空间对应的时空区域，构建所述训练集对应的时空区域集合；The training spatiotemporal region set generation module 404 is used to obtain the image feature space corresponding to the different action types, and based on the linear transformation method of random clustering forest, obtain the video feature space mapped by the image feature space, determine the spatiotemporal region corresponding to each of the video feature spaces, and construct the spatiotemporal region set corresponding to the training set;
测试时空区域确定模块405，用于将所述现场视频作为测试集，获取所述测试集中每一个动作对应的视频特征空间，确定所述视频特征空间对应的时空区域，作为测试时空区域；A test spatiotemporal region determination module 405 is used to use the live video as a test set, obtain a video feature space corresponding to each action in the test set, and determine a spatiotemporal region corresponding to the video feature space as a test spatiotemporal region;
舞蹈动作识别模块406，用于基于所述时空区域集合，确定所述测试时空区域对应的时空区域，基于所述时空区域识别所述现场视频中不同的舞蹈动作；A dance movement recognition module 406, configured to determine the space-time region corresponding to the test space-time region based on the space-time region set, and recognize different dance movements in the live video based on the space-time region;
舞蹈考试测评模块407，用于将所述现场视频中识别出舞蹈动作与预设的测试舞蹈动作进行对比，并基于预设算法确定两者间的相似度，将所述相似度作为测评结果，完成大学舞蹈考试测评。The dance test evaluation module 407 is used to compare the dance movements identified in the live video with the preset test dance movements, and determine the similarity between the two based on a preset algorithm, and use the similarity as the evaluation result to complete the university dance test evaluation.
本申请实施例所述的大学舞蹈考试测评装置，通过采集学生舞蹈考试的现场视频；构建舞蹈图片集；将舞蹈图片集传入预设的TLSVM模型，识别出不同动作种类；获取不同动作种类对应的时空区域，构建所述训练集对应的时空区域集合；获取现场视频中每一个动作对应的测试时空区域；从训练时空区域集合中确定测试时空区域一一对应的时空区域，基于时空区域识别现场视频中的舞蹈动作；将现场视频中识别出舞蹈动作与预设的测试舞蹈动作进行对比，确定两者间的相似度，将相似度作为测评结果，完成大学舞蹈考试测评。本申请避免监考老师的主观因素影响，做到公平进行舞蹈测评。The university dance test evaluation device described in the embodiment of the present application collects live videos of students' dance tests; constructs a dance picture set; passes the dance picture set into a preset TLSVM model to identify different types of movements; obtains the spatiotemporal regions corresponding to different types of movements, and constructs a set of spatiotemporal regions corresponding to the training set; obtains the test spatiotemporal regions corresponding to each movement in the live video; determines the spatiotemporal regions that correspond one to one to the test spatiotemporal regions from the training spatiotemporal region set, and identifies the dance movements in the live video based on the spatiotemporal regions; compares the dance movements identified in the live video with the preset test dance movements, determines the similarity between the two, and uses the similarity as the evaluation result to complete the university dance test evaluation. This application avoids the influence of subjective factors of the invigilator and ensures fair dance evaluation.
为解决上述技术问题，本申请实施例还提供计算机设备。具体请参阅图5，图5为本实施例计算机设备基本结构框图。To solve the above technical problems, the present application also provides a computer device. Please refer to FIG5 for details, which is a basic structural block diagram of the computer device of the present embodiment.
所述计算机设备5包括通过系统总线相互通信连接存储器5a、处理器5b、网络接口5c。需要指出的是，图中仅示出了具有组件5a-5c的计算机设备5，但是应理解的是，并不要求实施所有示出的组件，可以替代的实施更多或者更少的组件。其中，本技术领域技术人员可以理解，这里的计算机设备是一种能够按照事先设定或存储的指令，自动进行数值计算和/或信息处理的设备，其硬件包括但不限于微处理器、专用集成电路(ApplicationSpecific Integrated Circuit，ASIC)、可编程门阵列(Field-Programmable Gate Array，FPGA)、数字处理器(Digital Signal Processor，DSP)、嵌入式设备等。The computer device 5 includes a memory 5a, a processor 5b, and a network interface 5c that are interconnected and communicated through a system bus. It should be noted that the figure only shows a computer device 5 with components 5a-5c, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and/or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application-specific integrated circuits (ASIC), programmable gate arrays (FPGA), digital processors (DSP), embedded devices, etc.
所述计算机设备可以是桌上型计算机、笔记本、掌上电脑及云端服务器等计算设备。所述计算机设备可以与用户通过键盘、鼠标、遥控器、触摸板或声控设备等方式进行人机交互。The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.
所述存储器5a至少包括一种类型的可读存储介质，所述可读存储介质包括闪存、硬盘、多媒体卡、卡型存储器（例如，SD或DX存储器等）、随机访问存储器（RAM）、静态随机访问存储器（SRAM）、只读存储器（ROM）、电可擦除可编程只读存储器（EEPROM）、可编程只读存储器（PROM）、磁性存储器、磁盘、光盘等。在一些实施例中，所述存储器5a可以是所述计算机设备5的内部存储单元，例如该计算机设备5的硬盘或内存。在另一些实施例中，所述存储器5a也可以是所述计算机设备5的外部存储设备，例如该计算机设备5上配备的插接式硬盘，智能存储卡（Smart Media Card，SMC），安全数字（Secure Digital，SD）卡，闪存卡（FlashCard）等。当然，所述存储器5a还可以既包括所述计算机设备5的内部存储单元也包括其外部存储设备。本实施例中，所述存储器5a通常用于存储安装于所述计算机设备5的操作系统和各类应用软件，例如大学舞蹈考试测评方法的程序代码等。此外，所述存储器5a还可以用于暂时地存储已经输出或者将要输出的各类数据。The memory 5a includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 5a can be an internal storage unit of the computer device 5, such as a hard disk or memory of the computer device 5. In other embodiments, the memory 5a can also be an external storage device of the computer device 5, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. equipped on the computer device 5. Of course, the memory 5a can also include both the internal storage unit of the computer device 5 and its external storage device. In this embodiment, the memory 5a is generally used to store the operating system and various application software installed on the computer device 5, such as the program code of the college dance examination evaluation method, etc. In addition, the memory 5a can also be used to temporarily store various types of data that have been output or are to be output.
所述处理器5b在一些实施例中可以是中央处理器（Central Processing Unit，CPU）、控制器、微控制器、微处理器、或其他数据处理芯片。该处理器5b通常用于控制所述计算机设备5的总体操作。本实施例中，所述处理器5b用于运行所述存储器5a中存储的程序代码或者处理数据，例如运行所述大学舞蹈考试测评方法的程序代码。The processor 5b may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips in some embodiments. The processor 5b is generally used to control the overall operation of the computer device 5. In this embodiment, the processor 5b is used to run the program code stored in the memory 5a or process data, such as running the program code of the university dance test evaluation method.
所述网络接口5c可包括无线网络接口或有线网络接口，该网络接口5c通常用于在所述计算机设备5与其他电子设备之间建立通信连接。The network interface 5c may include a wireless network interface or a wired network interface. The network interface 5c is generally used to establish a communication connection between the computer device 5 and other electronic devices.
本申请还提供了另一种实施方式，即提供一种非易失性计算机可读存储介质，所述非易失性计算机可读存储介质存储有大学舞蹈考试测评程序，所述大学舞蹈考试测评程序可被至少一个处理器执行，以使所述至少一个处理器执行如上述的大学舞蹈考试测评方法的步骤。The present application also provides another embodiment, namely, providing a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a university dance examination and evaluation program, and the university dance examination and evaluation program can be executed by at least one processor so that the at least one processor performs the steps of the above-mentioned university dance examination and evaluation method.
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质（如ROM/RAM、磁碟、光盘）中，包括若干指令用以使得一台终端设备（可以是手机，计算机，服务器，空调器，或者网络设备等）执行本申请各个实施例所述的方法。Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present application.
显然，以上所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例，附图中给出了本申请的较佳实施例，但并不限制本申请的专利范围。本申请可以以许多不同的形式来实现，相反地，提供这些实施例的目的是使对本申请的公开内容的理解更加透彻全面。尽管参照前述实施例对本申请进行了详细的说明，对于本领域的技术人员来而言，其依然可以对前述各具体实施方式所记载的技术方案进行修改，或者对其中部分技术特征进行等效替换。凡是利用本申请说明书及附图内容所做的等效结构，直接或间接运用在其他相关的技术领域，均同理在本申请专利保护范围之内。Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to replace some of the technical features therein with equivalents. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011147588.5ACN112381118B (en) | 2020-10-23 | 2020-10-23 | A method and device for evaluating university dance examination |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011147588.5ACN112381118B (en) | 2020-10-23 | 2020-10-23 | A method and device for evaluating university dance examination |
| Publication Number | Publication Date |
|---|---|
| CN112381118A CN112381118A (en) | 2021-02-19 |
| CN112381118Btrue CN112381118B (en) | 2024-05-17 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011147588.5AActiveCN112381118B (en) | 2020-10-23 | 2020-10-23 | A method and device for evaluating university dance examination |
| Country | Link |
|---|---|
| CN (1) | CN112381118B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116612310B (en)* | 2023-07-17 | 2023-09-26 | 长春医学高等专科学校(长春职工医科大学长春市医学情报所) | Multimedia dance action based image decomposition processing method |
| CN117077084B (en)* | 2023-10-16 | 2024-01-26 | 南京栢拓视觉科技有限公司 | Dance scoring method based on space-time heterogeneous double-flow convolutional network |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799873A (en)* | 2012-07-23 | 2012-11-28 | 青岛科技大学 | Human body abnormal behavior recognition method |
| CN103310233A (en)* | 2013-06-28 | 2013-09-18 | 青岛科技大学 | Similarity mining method of similar behaviors between multiple views and behavior recognition method |
| CN108830252A (en)* | 2018-06-26 | 2018-11-16 | 哈尔滨工业大学 | A kind of convolutional neural networks human motion recognition method of amalgamation of global space-time characteristic |
| CN109508656A (en)* | 2018-10-29 | 2019-03-22 | 重庆中科云丛科技有限公司 | A kind of dancing grading automatic distinguishing method, system and computer readable storage medium |
| CN111563487A (en)* | 2020-07-14 | 2020-08-21 | 平安国际智慧城市科技股份有限公司 | Dance scoring method based on gesture recognition model and related equipment |
| KR20200119042A (en)* | 2019-04-09 | 2020-10-19 | 유진기술 주식회사 | Method and system for providing dance evaluation service |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799873A (en)* | 2012-07-23 | 2012-11-28 | 青岛科技大学 | Human body abnormal behavior recognition method |
| CN103310233A (en)* | 2013-06-28 | 2013-09-18 | 青岛科技大学 | Similarity mining method of similar behaviors between multiple views and behavior recognition method |
| CN108830252A (en)* | 2018-06-26 | 2018-11-16 | 哈尔滨工业大学 | A kind of convolutional neural networks human motion recognition method of amalgamation of global space-time characteristic |
| CN109508656A (en)* | 2018-10-29 | 2019-03-22 | 重庆中科云丛科技有限公司 | A kind of dancing grading automatic distinguishing method, system and computer readable storage medium |
| KR20200119042A (en)* | 2019-04-09 | 2020-10-19 | 유진기술 주식회사 | Method and system for providing dance evaluation service |
| CN111563487A (en)* | 2020-07-14 | 2020-08-21 | 平安国际智慧城市科技股份有限公司 | Dance scoring method based on gesture recognition model and related equipment |
| Title |
|---|
| 基于多目立体视觉的体育舞蹈错误动作校正方法;余海燕;;赤峰学院学报(自然科学版);20200725(第07期);全文* |
| Publication number | Publication date |
|---|---|
| CN112381118A (en) | 2021-02-19 |
| Publication | Publication Date | Title |
|---|---|---|
| CN111832468B (en) | Gesture recognition method and device based on biological recognition, computer equipment and medium | |
| CN111754541B (en) | Target tracking method, device, equipment and readable storage medium | |
| CN113139628B (en) | Sample image identification method, device and equipment and readable storage medium | |
| CN111680678B (en) | Target area identification method, device, equipment and readable storage medium | |
| CN110866469B (en) | Facial five sense organs identification method, device, equipment and medium | |
| WO2020107847A1 (en) | Bone point-based fall detection method and fall detection device therefor | |
| CN106874826A (en) | Face key point-tracking method and device | |
| JP7443647B2 (en) | Keypoint detection and model training method, apparatus, device, storage medium, and computer program | |
| CN110956131B (en) | Single-target tracking method, device and system | |
| CN114298997B (en) | Fake picture detection method, fake picture detection device and storage medium | |
| CN111507285A (en) | Face attribute recognition method and device, computer equipment and storage medium | |
| WO2022111387A1 (en) | Data processing method and related apparatus | |
| CN105096353A (en) | Image processing method and device | |
| CN112381118B (en) | A method and device for evaluating university dance examination | |
| CN112700462B (en) | Image segmentation method, device, electronic device and storage medium | |
| CN110163095B (en) | Loop detection method, loop detection device and terminal equipment | |
| CN119233047A (en) | Video sequence generation method, device, computer equipment and storage medium | |
| CN115861572B (en) | Three-dimensional modeling method, device, equipment and storage medium | |
| CN113822134B (en) | A video-based instance tracking method, device, equipment and storage medium | |
| CN113516735B (en) | Image processing method, device, computer readable medium and electronic device | |
| CN119228991A (en) | Method, device, equipment and medium for three-dimensional reconstruction of human face based on artificial intelligence | |
| CN110942056A (en) | Clothing key point positioning method and device, electronic equipment and medium | |
| CN112309181A (en) | Dance teaching auxiliary method and device | |
| CN115311498B (en) | Method, device, equipment and storage medium for training target detection model | |
| CN119293281B (en) | A model training method and related equipment for asymmetric image retrieval |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| OL01 | Intention to license declared | ||
| OL01 | Intention to license declared |