CN103390174A - Physical education assisting system and method based on human body posture recognition - Google Patents

Physical education assisting system and method based on human body posture recognition Download PDF

Info

Publication number
CN103390174A
CN103390174A CN 201210138503 CN201210138503A CN103390174A CN 103390174 A CN103390174 A CN 103390174A CN 201210138503 CN201210138503 CN 201210138503 CN 201210138503 A CN201210138503 A CN 201210138503A CN 103390174 A CN103390174 A CN 103390174A
Authority
CN
Grant status
Application
Patent type
Prior art keywords
standard
operation
user
action
time
Prior art date
Application number
CN 201210138503
Other languages
Chinese (zh)
Inventor
张琼
彭立焱
Original Assignee
深圳泰山在线科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date

Links

Abstract

The invention discloses a physical education assisting system and method based on human body posture recognition. The system comprises a depth shooting device, a standard action storage unit, a display terminal, an action comparison unit and a prompting device, wherein the depth shooting device is used for obtaining user action depth image sequences comprising movement action of a user following the demonstration action; the standard action storage unit is used for storing a standard action model and standard action demonstration image sequences corresponding to the standard action model; the display terminal is used for displaying the standard action demonstration image sequences for the user; the action comparison unit is used for comparing the user action depth image sequence at a scheduled time point or time period with the standard action model at the corresponding time point or the time period, and obtaining the difference between the user action and the standard action model; the prompting device is used for outputting action evaluation information or action correction prompting information to a display device according to the difference. According to the physical education assisting system and method, the user does not need to wear a special recognition device, the manufacturing cost of the system is lowered, and the user experience is improved.

Description

基于人体姿态识别的体育教学辅助系统和方法 Sports teaching aids and methods of human gesture recognition system based on

技术领域 FIELD

[0001] 本发明涉及计算机视觉技术,具体涉及基于人体姿态识别的体育教学辅助系统和方法。 [0001] The present invention relates to computer vision technology, particularly relates to a system and method for teaching aids PE-based human gesture recognition.

背景技术 Background technique

[0002] 传统体育教学需要场地和教练,受到训练者经济条件、训练场馆、训练时间等诸多限制,不利于大部分用户满足自己学习的目标。 [0002] Traditional physical education needs space and coach, trainer by economic conditions, training venues, training time, and many other restrictions, is not conducive to the majority of users to meet their learning goals. 借助教学视频学习没有互动性,也不能对自己的动作作出评价,在动作错误时也不能得到及时的纠正。 With instructional video learning is not interactive, nor to evaluate their own actions, but also in the operation of the error can not be corrected in time. 选择观看比赛或表演录像来通过模仿学习,虽然用户可以在任何时间任何环境完成,但是缺乏指导,对于动作没有评价体系,训练效果不能满足用户的需要。 Select the video to watch the game or performance learning through imitation, although the user can be done at any time and in any environment, but the lack of guidance for action no evaluation system, training effect can not meet the needs of users.

[0003] 现有技术中公开了采用佩戴人体关节标识装置并通过特定装置识别人体关节标识装置从而获取人体动作辅助体育教学的装置和对应的方法。 [0003] The prior art discloses a method of identifying apparatus using human body joints wear and thereby obtaining means and a corresponding operation of the auxiliary body through a specific sports teaching apparatus identifying means identifies human joints. 例如,中国专利申请CN102243687A公开了一种基于动作识别技术的体育教学辅助系统。 For example, Chinese patent application CN102243687A discloses a physical education support system based on motion recognition technology. 该系统通过为用户佩戴带有特定波长光源以及惯性测量单元的运动数据获取模块,由摄像机追踪所述特定波长光源同时采集惯性测量单元的测量结果来对目标进行动作识别,达到体育教学辅助的目的。 The system acquires data for the motion module with a user wears a specific wavelength of light source, and the inertial measurement unit to track the specific wavelength light by the camera while acquiring a measurement result of the inertial measurement unit is operated to identify the target, to achieve the purpose of teaching aids PE . 又例如,中国专利申请CN102000430A公开了一种基于计算机的舞蹈动作判定方法。 As another example, Chinese patent application CN102000430A discloses a computer-based method for judging dance movements. 该方法通过在人体设置跟踪点,采集跟踪点在人进行舞蹈时在不同时刻的空间位置,通过将跟踪点在不同时刻的位置进行投影计算其特征向量以对动作是否标准进行评价。 The method provided by the human body tracking point, the tracking point acquired at the time of person dancing spatial position different times, by projecting the position of the tracking point at different times to calculate the feature vectors to be evaluated whether the standard operation.

[0004] 但是,这类借助需要人体佩戴的识别装置的教学辅助系统,需要用户佩戴专用识别装置,虽然也能实现动作简单的体育运动的学习与比赛,但是不能用来学习动作变化比较多的体育运动,系统成本高。 [0004] However, such aid needs teaching assistant system identification device worn by the human body, require the user to wear special identification device, although it can achieve a simple action sports and learning the game, but can not be used to learn more action changes sports, high system costs.

[0005] 与此同时,人体姿态识别技术作为计算机视觉处理技术的分支,正在得到越来越广泛的应用。 [0005] At the same time, human gesture recognition technology as a branch of computer vision processing technology, it is being more widely used. 人体姿态识别技术通过对人体图像的数字处理判断人体姿态以及识别分割出图像中人体的不同部位的关节点,例如,头部和躯干等。 Body posture and gesture recognition techniques to identify the human body joints segmented images of different parts of the body, e.g., head and torso, etc. on the human body image by digital processing of judgment.

[0006]文献“Real-Time Human Pose Recognition in Parts from SingleDepth Images,,.Jamie Shotton, Andrew Fitzgibbon, Mat Cook, Toby Sharp, MarkFinocchio, Richard Moore, Alex Kipman, and Andrew Blake.CVPR, 2011 公开了一种深度图像实时人体姿态部分识别方法,通过对深度图像进行处理将困难的姿态估计问题转化为简单的逐像素分类问题实现了从深度图像中分离获取人体关节点坐标信息的目的。 [0006] Document "Real-Time Human Pose Recognition in Parts from SingleDepth Images ,,. Jamie Shotton, Andrew Fitzgibbon, Mat Cook, Toby Sharp, MarkFinocchio, Richard Moore, Alex Kipman, and Andrew Blake.CVPR, 2011 discloses a real-time human posture part depth image recognition method, by the depth image processing difficult pose estimation problem is transformed into a simple pixel-wise object classification achieved isolated from human joint depth image acquisition point coordinate information.

[0007] 综上,目前缺少亟需一种成本较低,不需要佩戴专门识别装置的体育教学辅助系统和方法。 The [0007] Fully, the current lack of a need for an low cost, does not need to wear PE Teaching System and method for uniquely identifying the device.

发明内容 SUMMARY

[0008] 本发明的目的在于提供一种无须佩戴识别点即可获取人体姿态信息进行体育教学辅导的装置和方法。 [0008] The object of the present invention is to provide an identification point to be worn without the body attitude information acquisition apparatus and method for tutoring PE.

[0009] 本发明公开了一种基于人体姿态识别的体育教学辅助系统,包括:[0010] 深度摄像装置,用于获取包括用户跟随示范动作运动动作的用户动作深度图像序列; [0009] The present invention discloses a Physical Teaching System Based on human gesture recognition, comprising: [0010] the depth of the imaging means for acquiring a user action comprising the user to follow the depth image of an exemplary operation sequence of the operation of the motion;

[0011] 标准动作存储单元,用于存储标准动作模型和与标准动作模型对应的标准动作示范图像序列; [0011] Normal operation of a storage unit for storing the standard operation of the exemplary image sequence standard motion model and motion model corresponding to the standard;

[0012] 显示终端,用于向用户显示所述标准动作示范图像序列; [0012] a display terminal for displaying the standard motion image sequence model to a user;

[0013] 动作比较单元,用于将预定时间点或时间段的所述用户动作深度图像序列与对应时间点或时间段的标准动作模型进行比较,获取用户动作与标准动作模型的差异; [0013] The operation of the comparison unit, the standard motion model point of time or a predetermined period of time a user operating point corresponding to the depth image sequence or the period of time for comparing, obtaining the difference with the standard operation of the user operation model;

[0014] 提示装置,用于根据所述差异向显示装置输出动作评价信息或动作纠正提示信肩、O [0014] presentation device, according to the difference to the display device outputs an operation to correct the operation of the evaluation information or prompts that shoulder, O

[0015] 其中,所述深度摄像装置还用于拍摄教练的标准动作获取标准动作深度图像序列,以及, [0015] wherein the depth of the imaging operation of the imaging means further coach for standard access to standard operation of the depth image sequences, and,

[0016] 所述系统还包括: The [0016] system further comprises:

[0017] 标准动作训练单元,用于根据所述标准动作深度图像序列建立标准动作模型; [0017] Standard training unit operation, to establish a standard motion model according to the standard sequence of operation of the depth image;

[0018] 普通摄像装置,用于获取标准动作示范图像序列,所述标准动作示范图像序列用于向用户示范标准动作; [0018] Ordinary image pickup means for acquiring the standard operation of an exemplary sequence of images, the standard sequence of operation of an exemplary standard image to a user operation model;

[0019] 其中,所述根据所述标准动作深度图像序列建立标准动作模型包括: [0019] wherein, said establishing comprises standard motion model according to the standard sequence of operation of the depth image:

[0020] 计算每帧标准动作深度图像的人体关节点的空间坐标构成人体姿态信息; [0020] Each spatial coordinate calculation standard action frame body joints depth image constituting the body posture information;

[0021] 根据连续多帧标准动作深度图像的人体姿态信息计算人体关节点的运动参数;其中,所述运动参数包括根据相邻帧的时间间隔和对应关节点的移动距离计算得到的关节点运动速度和人体肢体的转动角度; [0021] The continuous multi-standard operation information frame body posture calculating motion parameters depth image point of the joint body; wherein the motion parameter includes a motion joints calculated moving distance interval of temporally adjacent frames and the corresponding joint points speed and rotation angle of the human body;

[0022] 将所述人体关节点的人体姿态信息以及运动参数与图像序列时间信息一一映射获得标准动作模型。 [0022] The joints of the human body posture and motion parameter and the image information of the time sequence information obtained one mapping standard motion model.

[0023] 其中,所述将预定时间点或时间段的所述用户动作深度图像序列与对应时间点或时间段的标准动作模型进行比较,获取用户动作与标准动作模型的差异包括: [0023] wherein the standard model of the operation of the operating point of a user or a predetermined time period corresponding to the point sequence depth image or the time period is compared with the standard operation of the user acquired difference operation model comprises:

[0024] 计算预定时间点或时间段内每帧用户动作深度图像的人体关节点的空间坐标构成人体姿态信息; [0024] calculating a predetermined point in time or spatial coordinates of each user action frame body joints depth image period constitute the body posture information;

[0025] 根据连续多帧用户动作深度图像的人体姿态信息计算人体关节点的运动参数;其中,所述运动参数包括根据相邻帧的时间间隔和对应关节点的移动距离计算得到的关节点运动速度和人体肢体的转动角度; [0025] Human joints calculating motion parameters according to the continuous multi-user action frame body posture depth image information; wherein the motion parameter includes a moving distance calculated by the time interval of the adjacent frames and the corresponding points of articulation joint motions speed and rotation angle of the human body;

[0026] 将所述人体关节点的人体姿态信息以及运动参数与对应时间点或时间段的标准动作模型进行比较获取差异。 [0026] The operation of the standard model of the human body joints attitude information and motion parameters and the corresponding point of time or period of time compared acquires the difference.

[0027] 其中,所述系统还包括语音提示装置,所述提示装置输出通过语音信号提示用户纠正动作。 [0027] wherein said system further comprises a voice prompt means, means for outputting a speech signal by prompting the user to the prompt corrective action.

[0028] 其中,所述预定时间点为关键姿势的时间点,所述预定时间段为关键动作所处的时间段。 [0028] wherein said predetermined time point is a time point the key poses, the predetermined time period is a time period in which the operation key.

[0029] 本发明还公开了一种基于人体姿态识别的体育教学辅助方法,包括: [0029] The present invention also discloses a method of assisted teaching human PE-based gesture recognition, comprising:

[0030] 播放标准动作示范图像序列,同时通过深度摄像装置实时获取用户跟随所述标准动作示范图像序列运动的用户动作深度图像序列; [0030] Standard Play exemplary operation sequence of images, while obtaining a user operation of an exemplary image sequence to follow the movement of the user's operation of the standard depth of the depth image sequences in real-time imaging means;

[0031] 将预定时间点或时间段的所述用户动作深度图像序列与对应时间点或时间段的标准动作模型进行比较,获取用户动作与标准动作模型的差异; [0031] The operation of the standard model of the user or the operating point of the predetermined time period corresponding to the depth of the image point in time or sequence period are compared, obtaining the difference with the standard operation of the user operation model;

[0032] 根据所述差异评价用户动作或给出动作纠正提示。 [0032] The difference between the evaluation given by a user action or prompt corrective action.

[0033] 优选地,所述方法还包括: [0033] Preferably, the method further comprising:

[0034] 通过深度摄像装置和普通摄像装置分别获取标准动作深度图像序列和标准动作示范图像序列; [0034] The operation of the depth image obtaining standard sequences and the standard operation of an exemplary sequence of images by the imaging device and the depth of ordinary image pickup devices, respectively;

[0035] 根据所述标准动作深度图像序列建立标准动作模型。 [0035] The depth image sequences establish a standard motion model according to the standard operation.

[0036] 本发明能显著提高用户学习体育运动的效率,寓教于乐,通过语音和视频享受到体育中乐趣,本发明特有的互动性能让学习者享受到如同专业教练陪练的体验,能够提供给学习者更加准确规范的动作,学习性与娱乐性兼具。 [0036] The present invention can significantly improve the efficiency of the user to learn the sport, entertaining, enjoy the fun of sports through voice and video, unique interactive performance of the present invention allows the learner to enjoy sparring experience as a professional coach, can provide learners more accurate standardized operation of both learning and entertainment. 同时,由于不需要佩戴标记点等专门用具,本发明降低了系统成本同时提升了用户体验。 Meanwhile, there is no need to wear special appliances other markers, the present invention reduces the system costs while improving user experience.

附图说明 BRIEF DESCRIPTION

[0037]图1是本发明第一实施例的体育教学辅助系统的系统框图; [0037] FIG. 1 is a system block diagram of PE teaching assistance system according to a first embodiment of the present invention;

[0038]图2是本发明第二实施例的体育教学辅助系统的系统框图; [0038] FIG. 2 is a system block diagram of a second sports teaching assistance system of the present embodiment of the invention;

[0039] 图3是本发明实施例中根据深度图像获取表示人体姿态的关节点的示意图; [0039] FIG. 3 is a schematic diagram showing joints obtaining the body posture of the depth image according to embodiments of the present invention;

[0040] 图4是本发明实施例的体育教学辅助方法的方法流程图。 [0040] FIG. 4 is a method of assisted teaching methods PHYSICAL embodiment of the present invention a flowchart.

具体实施方式 detailed description

[0041] 下面结合附图并通过具体实施方式来进一步说明本发明的技术方案。 [0041] below with reference to specific embodiments and further technical solutions of the present invention.

[0042]图1是本发明第一实施例的体育教学辅助系统的系统框图。 [0042] FIG. 1 is a system block diagram of PE teaching assistance system according to a first embodiment of the present invention.

[0043] 如图1所示,本实施例的体育教学辅助系统包括:深度摄像装置101 (DepthCamera,也称为3D摄像装置,可以获取包括物体距离摄像装置的距离信息,因此可以基于此获得拍摄物体的三维空间坐标)、标准动作存储单元102、动作比较单元103、提示装置105和显示装置104。 [0043] As shown in FIG 1, PE teaching aid system according to the present embodiment includes: a depth image pickup apparatus 101 (DepthCamera, also referred to as 3D imaging device may acquire the distance information including the object distance of the imaging device, imaging can be obtained based on this three-dimensional coordinates of the object), the operation of the standard storage unit 102, the operation of the comparison unit 103, indicating means 105, and a display device 104.

[0044] 其中,所述深度摄像装置101用于获取包括用户跟随示范动作运动动作的用户动作深度图像序列。 [0044] wherein the depth of the imaging device 101 for acquiring a user action comprising the user to follow the depth image sequences exemplary operation of the moving operation. 用户动作深度图像序列用于来与标准动作模型进行比较以获得用户动作与标准动作之间的差异。 Sequence of user actions for the depth image to compare with the standard motion model to obtain the difference between the standard operation of a user action.

[0045] 所述标准动作存储单元102存储预先制作的用于进行动作比较的标准动作模型和用于显示给用户做示范用的标准动作示范图像序列。 The [0045] storage unit 102 stores operation standard prepared in advance for the standard motion model and a comparison operation for display to the user as the standard sequence of operation of an exemplary image with exemplary. 所述标准动作动作模型可以通过拍摄标准动作获取标准动作深度图像序列并进一步对图像序列进行训练获得。 The standard action motion models may be obtained by standard sequence of operation of the depth image capturing operation and further the standard training sequence of images is obtained. 所述标准动作存储单元102向动作比较单元103提供所述标准动作模型,向显示装置105提供标准动作示范图像序列供用户跟随学习。 The standard operation of the memory unit 102 to provide the standard motion model operation comparison unit 103, 105 provides a standard sequence of operation of the exemplary image display apparatus for the user to follow the study.

[0046] 本实施例中,标准动作存储单元102可以采用现有的各类存储装置,如只读存储器、可读写存储器、硬盘、闪存、光盘等,事先将制作好的相关标准动作数据烧录/存储到标准动作存储单元中即可。 [0046] In this embodiment, the standard operation of the memory unit 102 may employ various types of conventional memory devices such as read only memory, a readable and writable memory, a hard disk, a flash memory, an optical disk, a good production standards beforehand burn operation data recorded / stored in the storage unit to the standard operation.

[0047] 动作比较单元103用于将预定时间点或时间段的所述用户动作深度图像序列解析为用户姿态信息,将用户姿态信息与对应时间点或时间段的标准动作模型进行比较,获取用户动作与标准动作模型的差异。 [0047] The operation of the comparison unit 103 for a predetermined time or user action time point depth image sequence analysis for the user gesture information, standard user gesture motion model information and the corresponding point of time or time period are compared to obtain the user differences with the standard action model of action. 所述的预定时间点或时间段是事先指定的关键姿势所处时间点和关键动作的时间段,例如,广播体操开始时人体的预备姿势以及广播体操的重要动作。 Said predetermined point of time or time period is a time period designated in advance and the key poses time critical action point is located, e.g., gymnastics broadcast start preliminary operation of the body posture, and an important broadcast gymnastics. 动作比较单元103仅对该关键姿势和关键动作来进行模型比较,以获得差异,并进一步由提示装置来进行评价和纠正。 Only the operation of the comparison unit 103 and a key operation of key poses are compared to the model, in order to obtain the difference, and is further evaluated and corrected by the presentation device. 获取的差异包括获取用户姿态信息与标准动作模型的关节点之间的距离,肢体转动角度和关节点运动速度的差异。 Obtaining includes obtaining the difference between the distance from the joint point of the user with the standard operation of the posture information model, limb joints and the rotation angle of the difference in velocity.

[0048] 提示装置104用于根据所述差异评价用户动作或给出动作纠正提示。 [0048] The presentation device 104 according to the difference between the evaluation given by a user action or prompt corrective action. 提示装置106根据用户姿态信息与标准动作模型的关节点距离,肢体转动角度和关节点运动速度等差异按预定的评价标准对用户动作评价,还可以根据所述差异给出纠正动作的提示,提示装置106将评价信息或提示信息输出到显示装置显示。 The joint device 106 prompts the user points the standard attitude information from the operation model, the rotational angle differences in limb joints and the like to the user operation velocity evaluated according to a predetermined evaluation standard, can also prompt corrective action based on the difference, suggesting evaluation means 106 output information or prompt information to the display device displays. 例如,对于广播体操中平展手臂的姿势,动作比较单元105获取的用户动作图像与标准动作模型的差异为人体手部关节点和肘部关节点的坐标低于标准动作模型中的相应坐标,则判断用户手臂偏低,提示装置106提示用户手臂应当抬高。 User operation of the image difference operation with the standard model example, for spreading broadcast gymnastics arm posture, the comparison unit 105 acquires the operation of the human hand joints and elbow joints is lower than the corresponding coordinate of the coordinates of the standard model of operation, the Analyzing the user's arm is low, device 106 prompts the user to prompt the arm should be raised. 为了克服人体体型不同的问题,可以在获取人体姿态信息时对于关节点的坐标进行归一化。 To overcome the problem of different body types, can be normalized to coordinate joint point in obtaining the body posture information.

[0049] 显示装置105用于向用户显示所述标准动作示范图像序列。 [0049] The display means 105 for displaying the standard motion image sequence model to the user.

[0050]图2是本发明第二实施例的体育教学辅助系统的系统框图。 [0050] FIG. 2 is a system block diagram of a second sports teaching assistance system of the present embodiment of the invention. 本发明的第二实施例在第一所实施例基础上增加了普通摄像装置以及标准动作训练单元从而使得所述教学辅助系统具备制作标准动作数据的功能。 The second embodiment of the present invention in a first embodiment based on the increase of the ordinary operation of the imaging device and a standard training unit so that the secondary system includes a calibration instruction data operation functions.

[0051] 如图2所示,本实施例的体育教学辅助系统包括:深度摄像装置201、普通摄像装置202 (二维摄像装置,能够获取具有灰度信息或彩色信息的二维图像)、显示装置203、标准动作训练单元204、动作比较单元205、提示装置206和标准动作存储单元207。 [0051] 2, PE teaching aid system according to the present embodiment includes: a depth image pickup device 201, an ordinary image pickup device 202 (two-dimensional image pickup apparatus capable of acquiring a two-dimensional image information having a gray scale or color information), a display device 203, a standard training operation unit 204, the operation of the comparison unit 205, indicating means 206 and the standard operation of the memory unit 207.

[0052] 其中,深度摄像装置201用于获取标准动作深度图像序列和包括用户跟随示范动作运动动作的用户动作深度图像序列。 [0052] wherein the depth of the imaging apparatus 201 for obtaining the standard operation of the depth image sequences and comprising a user to follow the depth image sequence of user actions exemplary operation of the moving operation. 标准动作深度图像序列通过拍摄教练的标准动作录制,用来供标准动作训练单元训练获取标准动作模型。 Standard Action depth image sequences recorded by standard action shooting coach, used for the standard action training modules of training to obtain a standard action model. 用户动作深度图像序列用于来与标准动作模型进行比较以获得用户动作与标准动作之间的差异。 Sequence of user actions for the depth image to compare with the standard motion model to obtain the difference between the standard operation of a user action.

[0053] 普通摄像装置202与上述深度摄像装置201同步拍摄教练的标准动作获取标准动作示范图像序列,所述标准动作示范图像序列用于向用户示范标准动作。 [0053] The image pickup device 202 and the above-described general depth imaging apparatus 201 imaging synchronizing the operation of access standards standard coach exemplary operation sequence of images, said sequence of images for an exemplary operation of the standard user exemplary standard action.

[0054] 显示装置203用于向用户显示所述标准动作示范图像序列。 [0054] The display device 203 for displaying the standard motion image sequence model to the user.

[0055] 标准动作训练单元204用于根据所述标准动作深度图像序列建立标准动作模型。 [0055] Standard operation training unit 204 to establish a standard motion model according to the standard sequence of operation of the depth image.

[0056] 其中,建立标准动作模型的流程包括,标准动作训练单元204计算每巾贞标准动作深度图像的人体关节点的空间坐标构成人体姿态信息;根据连续多帧标准动作深度图像的人体姿态信息计算人体关节点的运动参数;将所述人体关节点的人体姿态信息以及运动参数与图像序列时间信息一一映射获得标准动作模型。 [0056] wherein, the establishment of a standard motion model of the process includes a standard movement training unit 204 calculates the spatial coordinates of the body joints of each towel Zhen standard action depth image constituting the body posture information; according to the continuous multi-frame body posture standard action depth image computing motion parameters joints of the human body; joints of the human body posture and motion parameter and the image information of the time sequence information obtained one mapping standard motion model.

[0057] 图3是从人体深度图像中提取人体关节点表示人体姿态的示意图。 [0057] FIG. 3 is extracted from the human body joints depth image schematic representation body posture. 如图3所示,利用一系列的人体关节点即可以表示人体的姿态。 As shown in FIG 3, using a series of joints of the human body which can represent the gesture. 利用所述的一系列的关节点的坐标即可构成人体姿态信息。 Using a series of coordinates of the joints to constitute the body posture information.

[0058] 静态的姿势利用单帧深度图像即可以构建。 [0058] static posture i.e. the depth image using a single frame can be constructed. 而对于动态的动作,可以将动作分解成若干个姿态,不同姿态下关节点的空间变化可以反映动作完成情况,根据动作期间不同关节点三维信息的变化特征建立包括动作特征的模型。 For dynamic operation, the operation may be divided into several posture, spatial variation of different joints gesture may reflect the completion of the operation, the action model includes three-dimensional features in accordance with different variation characteristic point information during operation of the joint. 在本实施例中,根据连续多帧的标准动作深度图像的人体姿态信息计算人体关节点的运动参数。 In the present embodiment, the motion parameters calculated according to the body joints of the standard operation of the continuous multi-frame body posture information of the depth image. 运动参数可以包括关节点运动的速度以及人体肢体转动的角度。 Motion parameters may include speed and angle of rotation of the limb joints of human motion. 所述速度可以通过相邻帧的间隔时间以及相邻帧的图像中关键点移动的距离计算,人体肢体转动的角度可以通过关节点构成的代表肢体的线段夹角计算得到。 The speed by the time interval and the distance between adjacent frames in an image frame adjacent keys mobile computing, on behalf of the angle limb segment human body may constitute the rotation angle calculated by joints.

[0059] 以广播体操为例,对于广播体操中的手臂上扬动作,标准动作训练单元104根据整个手臂上扬动作期间的多帧图像获得每一帧图像的人体姿态信息,通过连续的人体姿态信息,计算每一帧中对应的人体关节点的位置、移动速度以及所有肢体角度,然后将这些信息与对应的时间信息进行映射构成标准动作模型,该标准动作模型中包括手臂上扬动作中手臂对应的关节点的变化规律。 [0059] In an example broadcast gymnastics, gymnastics for broadcasting arm up operation, standard operation training unit 104 obtains each frame image according to the multi-frame image during the operation of the entire arm body posture information up, by a continuous body posture information, computing the human joints each frame corresponding to a position, moving speed and angle of the limb in all, the time information and information constituting the map corresponding to the standard motion model, the model includes a standard action arm up operation corresponding to the arm joint variation points.

[0060] 动作比较单元205用于将预定时间点或时间段的所述用户动作深度图像序列解析为用户姿态信息,将用户姿态信息与对应时间点或时间段的标准动作模型进行比较,获取用户动作与标准动作模型的差异。 [0060] The operation of the comparison unit 205 for a predetermined time or user action time point depth image sequence analysis for the user gesture information, standard user gesture motion model information and the corresponding point of time or time period are compared to obtain the user differences with the standard action model of action. 所述的预定时间点或时间段是事先指定的关键姿势所处时间点和关键动作的时间段,例如,广播体操开始时人体的预备姿势以及广播体操的重要动作。 Said predetermined point of time or time period is a time period designated in advance and the key poses time critical action point is located, e.g., gymnastics broadcast start preliminary operation of the body posture, and an important broadcast gymnastics. 动作比较单元105仅对该关键姿势和关键动作来进行模型比较,以获得差异,并进一步由提示装置来进行评价和纠正。 Only the operation of the comparison unit 105 and a key operation of key poses are compared to the model, in order to obtain the difference, and is further evaluated and corrected by the presentation device. 获取的差异包括获取用户姿态信息与标准动作模型的关节点距离,动作转动角度和关节点运动速度的差异。 Difference acquiring includes acquiring the posture information joints user model from the standard operation, the operation angle and the difference in rotational speed of movement of the joints.

[0061] 提示装置206用于根据所述差异评价用户动作或给出动作纠正提示。 [0061] The presentation device 206 according to the difference between the evaluation given by a user action or prompt corrective action. 提示装置206根据用户姿态信息与标准动作模型的关节点距离,动作转动角度和关节点运动速度等差异按预定的评价标准对用户动作评价,还可以根据所述差异给出纠正动作的提示,提示装置206将评价信息或提示信息输出到显示装置显示。 The joint device 206 prompts the user points the standard attitude information from the operation model, the operation of the rotational angle velocity joints and other user actions difference evaluated according to a predetermined evaluation standard, can also prompt corrective action based on the difference, suggesting evaluation means 206 output information or prompt information to the display device displays. 例如,对于广播体操中平展手臂的姿势,动作比较单元205获取的用户动作图像与标准动作模型的差异为人体手部关节点和肘部关节点的坐标低于标准动作模型中的相应坐标,则判断用户手臂偏低,提示装置206提示用户手臂应当抬高。 User operation of the image difference operation with the standard model example, for spreading broadcast gymnastics arm posture, the comparison unit 205 acquires the operation of the human hand joints and elbow joints is lower than the corresponding coordinate of the coordinates of the standard model of operation, the Analyzing the user's arm is low, device 206 prompts the user to prompt the arm should be raised. 为了克服人体体型不同的问题,可以在获取人体姿态信息时对于关节点的坐标进行归一化。 To overcome the problem of different body types, can be normalized to coordinate joint point in obtaining the body posture information.

[0062] 标准动作存储单元207用于存储标准动作训练单元204获得的标准动作模型以及由普通摄像装置202获取的标准动作示范图像序列。 Standard motion models [0062] Normal operation of the memory unit 207 for storing a normal operation, and the training unit 204 obtains the standard sequence of operation of an exemplary image acquired by ordinary image pickup device 202.

[0063] 在本发明的另一个`实施例中,`基于人体姿态识别的体育教学辅助系统还可以包括音频输出装置,所述提示装置206同时将评价信息或提示信息以语音方式输出到音频输出 [0063] In another 'embodiment of the present invention, the body `PE Teaching System based gesture recognition may further include an audio output device, the presentation device 206 while the evaluation information or prompt information to the audio output by voice output

>jU ρςα装直。 > JU ρςα loaded straight.

[0064] 同时,由于用户是跟随示范动作在进行,用户动作相比于标准动作时间上会有一定的滞后,在系统处理用户的每个动作时,会允许一定的时间偏差量,以保持进行比较时,用户和教练的视频同步。 [0064] Meanwhile, since the user is performing the operation following the demonstration, the user action as compared to the standard operation there will be some lag time, at each operation of system processes the user, will allow a certain amount of time offsets, to be held when compared to the user and coaches video synchronization.

[0065] 图4是本发明实施例的体育教学辅助方法的方法流程图。 [0065] FIG. 4 is a method of assisted teaching methods PHYSICAL embodiment of the present invention a flowchart. 如图4所示,所述方法包括: , Shown in Figure 4, the method comprising:

[0066] 步骤401、播放预先制作的标准动作示范图像序列,同时通过深度摄像装置实时获取用户跟随所述标准动作示范图像序列运动的用户动作深度图像序列; [0066] Step 401, the standard playback operation of an exemplary sequence of images prepared in advance, while exemplary operation of obtaining the user follows the user of the motion image sequence standard action depth image sequences in real time depth imaging means;

[0067] 步骤4 O 2、将预定时间点或时间段的所述用户动作深度图像序列与对应时间点或时间段的标准动作模型进行比较,获取用户动作与标准动作模型的差异; [0067] Step 4 O 2, the point of the standard motion model user action point or a predetermined time period corresponding to the depth image sequence or the period of time compared with the standard operation of a user acquires difference motion model;

[0068] 步骤403、根据所述差异评价用户动作或给出动作纠正提示。 [0068] Step 403, based on the difference evaluation given by user action or prompt corrective action.

[0069] 其中,步骤402进一步包括: [0069] wherein, the step 402 further comprises:

[0070] 402Α、计算预定时间点或时间段内每帧用户动作深度图像的人体关节点的空间坐标构成人体姿态信息;[0071] 402B、根据连续多帧用户动作深度图像的人体姿态信息计算人体关节点的运动参数;其中,所述运动参数包括根据相邻帧的时间间隔和对应关节点的移动距离计算得到的关节点运动速度和人体肢体的转动角度; [0070] 402Α, spatial coordinates of each user action frame body joints depth image calculation point of time or a predetermined period of time constitute the body posture information; [0071] 402B, is calculated according to the body frame body posture information of the continuous multi-user action depth image motion parameters articulation point; wherein the motion parameter includes a motion velocity joints and the rotation angle of the human limb movement distance calculated based on the time interval of the adjacent frames and the corresponding point of the joint;

[0072] 402C、将所述人体关节点的人体姿态信息以及运动参数与对应时间点或时间段的标准动作模型进行比较获取差异。 [0072] 402C, a standard motion model of the human body joints and body posture information of the motion parameters and the corresponding point of time or period of time compared acquires the difference.

[0073] 其中,在步骤401前还可包括制作获取标准动作示范图像序列和获取标准动作模型的步骤,具体包括: [0073] wherein, in the step 401 may further include a production prior to obtaining the standard operation of an exemplary standard image sequence acquisition step and motion models comprises:

[0074] 400A、分别获取标准动作深度图像序列和标准动作示范图像序列。 [0074] 400A, respectively, obtain the standard sequence of operation of the depth image and the standard operation of the exemplary image sequence.

[0075] 400B、根据所述标准动作深度图像序列建立标准动作模型。 [0075] 400B, depth image sequences establish a standard motion model according to the standard operation.

[0076] 其中,所述步骤400B进一步包括: [0076] wherein said step 400B further comprises:

[0077] 步骤400B1、计算每巾贞标准动作深度图像的人体关节点的空间坐标构成人体姿态 Spatial coordinates [0077] Step 400B1, Chen calculated for each standard action towel depth image points constituting human joint body posture

信息; information;

[0078] 步骤400B2、根据连续多帧标准动作深度图像的人体姿态信息计算人体关节点的运动参数; [0078] Step 400B2, is calculated according to the body joints of the continuous multi-frame body posture information of the depth image of the standard operation of motion parameters;

[0079] 步骤400B3、将所述人体关节点的人体姿态信息以及运动参数与图像序列时间信息映射获得标准动作模型。 [0079] Step 400B3, the joints of the human body posture parameter and the image information and the motion information map obtained standard time sequence motion model.

[0080] 本发明能显著提高用户学习体育运动的效率,寓教于乐,通过语音和视频享受到体育中乐趣,本发明特有的互动性能让学习者享受到如同专业教练陪练的体验,能够提供给学习者更加准确规范的动作,学习性与娱乐性兼具。 [0080] The present invention can significantly improve the efficiency of the user to learn the sport, entertaining, enjoy the fun of sports through voice and video, unique interactive performance of the present invention allows the learner to enjoy sparring experience as a professional coach, can provide learners more accurate standardized operation of both learning and entertainment. 同时,由于不需要佩戴标记点等专门用具,本发明降低了系统成本同时提升了用户体验。 Meanwhile, there is no need to wear special appliances other markers, the present invention reduces the system costs while improving user experience.

[0081] 显然,本领域技术人员应该明白,上述的本发明的各模块或各步骤可以用通用的计算装置来实现,它们可以集中在单个计算装置上,或者分布在多个计算装置所组成的网络上,可选地,他们可以用计算机装置可执行的程序代码来实现,从而可以将它们存储在存储装置中由计算装置来执行,或者将它们分别制作成各个集成电路模块,或者将它们中的多个模块或步骤制作成单个集成电路模块来实现。 [0081] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be realized with general computing devices, they can be integrated on a single computing device or distributed across multiple computing devices consisting of network, program codes, they may be executable by the computer device implemented thereby may be performed by a computing device stored in a storage device, or they are made into integrated circuit modules, or they are a plurality of modules or steps are manufactured into a single integrated circuit module. 这样,本发明不限制于任何特定的硬件和软件的结合。 Thus, the present invention is not limited to any specific combination of hardware and software.

[0082] 以上所述仅为本发明的优选实施例,并不用于限制本发明,对于本领域技术人员而言,本发明可以有各种改动和变化。 [0082] The above are only preferred embodiments of the present invention is not intended to limit the invention to those skilled in the art, the present invention may have various modifications and changes. 凡在本发明的精神和原理之内所作的任何修改、等同替换、改进等,均应包含在本发明的保护范围之内。 Any modifications within the spirit and principle of the invention, equivalent substitutions, improvements, etc., should be included within the scope of the present invention.

Claims (12)

  1. 1.一种基于人体姿态识别的体育教学辅助系统,包括: 深度摄像装置,用于获取包括用户跟随示范动作运动动作的用户动作深度图像序列; 标准动作存储单元,用于存储标准动作模型和与标准动作模型对应的标准动作示范图像序列; 显示终端,用于向用户显示所述标准动作示范图像序列; 动作比较单元,用于将预定时间点或时间段的所述用户动作深度图像序列与对应时间点或时间段的标准动作模型进行比较,获取用户动作与标准动作模型的差异; 提示装置,用于根据所述差异向显示装置输出动作评价信息或动作纠正提示信息。 A teaching aid system based on Physical body gesture recognition, comprising: a depth image pickup means for acquiring a user action comprising the user to follow the depth image sequences exemplary operation of the moving operation; standard operation of a storage unit for storing a standard action and a model standard exemplary operation sequence of the image corresponding to the standard motion model; display terminal for displaying the standard motion image sequence model to a user; comparing unit operation, the user operating point for a predetermined period of time or a sequence of images corresponding to the depth standard motion model point in time or time period are compared with the standard operation of a user acquires difference motion models; prompt means, means for outputting an operation or action to correct the evaluation information according to the difference message to the display.
  2. 2.如权利要求1所述的基于人体姿态识别的体育教学辅助系统,其特征在于,所述深度摄像装置还用于拍摄教练的标准动作获取标准动作深度图像序列,以及, 所述系统还包括: 标准动作训练单元,用于根据所述标准动作深度图像序列建立标准动作模型; 普通摄像装置,用于获取标准动作示范图像序列,所述标准动作示范图像序列用于向用户示范标准动作。 2. Physical Teaching System based on human gesture recognition, wherein the 1, said apparatus further depth camera for photographing the standard operation of the standard operation of acquiring coach depth image sequences, and, the system according to claim further comprising : standard action training unit, according to the criteria established for the operation of the depth image sequence standard motion model; normal imaging means for acquiring the standard operation of an exemplary sequence of images, the standard sequence of operation of exemplary images used to demonstrate the operation of the standard to the user.
  3. 3.如权利要求2所述的基于人体姿态识别的体育教学辅助系统,其特征在于,所述根据所述标准动作深度图像序列建立标准动作模型包括: 计算每帧标准动作深度图像的人体关节点的空间坐标构成人体姿态信息; 根据连续多帧标准动作深度图像的人体姿态信息计算人体关节点的运动参数;其中,所述运动参数包括根据相邻帧的时间间隔和对应关节点的移动距离计算得到的关节点运动速度和人体肢体的转动角度; 将所述人体关节点的人体姿态信息以及运动参数与图像序列时间信息一一映射获得标准动作模型。 Each calculation operation frame body joints standard depth image: Teaching System Physical body based gesture recognition, wherein the 2, including the establishment of a standard motion model according to the standard sequence of operation of the depth image as claimed in claim the spatial coordinates constitute the body posture information; calculating motion parameters according to body joints in the continuous multi-standard operation of the frame body posture information of the depth image; wherein the motion parameter includes calculating a moving distance interval of temporally adjacent frames and the corresponding joint points velocity joints and the rotation angle obtained human limb; the joints of the human body posture and motion parameter and the image information of the time sequence information obtained one mapping standard motion model.
  4. 4.如权利要求3所述的基于人体姿态识别的体育教学辅助系统,其特征在于,所述将预定时间点或时间段的所述用户动作深度图像序列与对应时间点或时间段的标准动作模型进行比较,获取用户动作与标准动作模型的差异包括: 计算预定时间点或时间段内每帧用户动作深度图像的人体关节点的空间坐标构成人体姿态信息; 根据连续多帧用户动作深度图像的人体姿态信息计算人体关节点的运动参数;其中,所述运动参数包括根据相邻帧的时间间隔和对应关节点的移动距离计算得到的关节点运动速度和人体肢体的转动角度; 将所述人体关节点的人体姿态信息以及运动参数与对应时间点或时间段的标准动作模型进行比较获取差异。 4. Physical Teaching System based on human gesture recognition, wherein said 3, the standard operation of the operating point of the user or a predetermined time period and the depth image corresponding to a sequence of time points or time period claims comparing the model, obtaining the difference with the standard operation of the user operation model comprises: calculating a predetermined point of time or period of operation of the user space coordinates of each frame body joints depth image information constituting the body posture; according to the continuous multi-user action depth image frames body posture information calculating motion parameters joints of the body; wherein the motion parameter includes a motion velocity joints and the rotation angle of the body limb moving distance interval of temporally adjacent frames and the corresponding joint points calculated; the human joints and the body posture information of the standard motion model and motion parameters corresponding to a time point or time period compared acquires the difference.
  5. 5.如权利要求1所述的基于人体姿态识别的体育教学辅助系统,其特征在于,所述系统还包括语音提示装置,所述提示装置输出通过语音信号提示用户纠正动作。 5. Physical Teaching System based on human gesture recognition, wherein the 1, said system further comprising a voice prompt means, said output means by suggesting a voice signal to prompt the user to correct the operation of the claims.
  6. 6.如权利要求1所述的基于人体姿态识别的体育教学辅助系统,其特征在于,所述预定时间点为关键姿势的时间点,所述预定时间段为关键动作所处的时间段。 6. Physical Teaching System based on human gesture recognition, wherein the 1, wherein the predetermined time point as the time point of key poses, the predetermined time period is a time period in which the operation key claims.
  7. 7.一种基于人体姿态识别的体育教学辅助方法,包括: 播放标准动作示范图像序列,同时通过深度摄像装置实时获取用户跟随所述标准动作示范图像序列运动的用户动作深度图像序列;将预定时间点或时间段的所述用户动作深度图像序列与对应时间点或时间段的标准动作模型进行比较,获取用户动作与标准动作模型的差异; 根据所述差异评价用户动作或给出动作纠正提示。 A secondary PE teaching method based on the human gesture recognition, comprising: a playback operation of an exemplary standard image sequence, while a depth image obtaining operation of the user to follow the normal sequence of operation of the user exemplary sequence moving image in real time by the depth camera means; predetermined time standard motion model of the user action point or points of time corresponding to the depth image sequence or the period of time is compared, the difference obtaining user operation of the standard motion model; evaluated according to the difference given by user action or prompt corrective action.
  8. 8.如权利要求7所述的基于人体姿态识别的体育教学辅助方法,其特征在于,所述方法还包括: 通过深度摄像装置和普通摄像装置分别获取标准动作深度图像序列和标准动作示范图像序列; 根据所述标准动作深度图像序列建立标准动作模型。 Obtaining standard sequence of operation of the depth image and the standard image sequence by operation of the exemplary depth camera means and image pickup means are common: 8. Physical assisted teaching methods based on human gesture recognition, wherein said 7, said method further including the claims ; depth image sequences establish a standard motion model according to the standard operation.
  9. 9.如权利要求8所述的基于人体姿态识别的体育教学辅助方法,其特征在于,所述根据所述标准动作深度图像序列建立标准动作模型包括: 计算每帧标准动作深度图像的人体关节点的空间坐标构成人体姿态信息; 根据连续多帧标准动作深度图像的人体姿态信息计算人体关节点的运动参数;所述运动参数包括根据相邻帧的时间间隔和对应关节点的移动距离计算得到的关节点运动速度和人体肢体的转动角度; 将所述人体关节点的人体姿态信息以及运动参数与图像序列时间信息一一映射获得标准动作模型。 Each calculation operation frame body joints standard depth image: Physical assisted teaching methods based on human gesture recognition, wherein said 8, comprising the establishment of a standard motion model according to the standard sequence of operation of the depth image according to claim the spatial coordinates constitute the body posture information; calculating motion parameters according to body joints in the continuous multi-standard operation of the frame body posture information of the depth image; the motion parameter includes a moving distance and the time interval corresponding to adjacent frames articulation points calculated velocity joints and the rotation angle of the human body; joints of the human body posture and motion parameter and the image information of the time sequence information obtained one mapping standard motion model.
  10. 10.如权利要求7所述的基于人体姿态识别的体育教学辅助方法,其特征在于,所述将预定时间点或时间段的所述用户动作深度图像序列与对应时间点或时间段的标准动作模型进行比较,获取用户动作与标准动作模型的差异包括: 计算预定时间点或时间段内每帧用户动作深度图像的人体关节点的空间坐标构成人体姿态信息; 根据连续多帧用户动作深度图像的人体姿态信息计算人体关节点的运动参数;其中,所述运动参数包括根据相邻帧的时间间隔和对应关节点的移动距离计算得到的关节点运动速度和人体肢体的转动角度; 将所述人体关节点的人体姿态信息以及运动参数与对应时间点或时间段的标准动作模型进行比较获取差异。 Physical teaching assistance method based on human gesture recognition, wherein said 7, the standard operation of the user or the operating point of the predetermined period of time points corresponding to the depth image sequence or the period of time according to claim comparing the model, obtaining the difference with the standard operation of the user operation model comprises: calculating a predetermined point of time or period of operation of the user space coordinates of each frame body joints depth image information constituting the body posture; according to the continuous multi-user action depth image frames body posture information calculating motion parameters joints of the body; wherein the motion parameter includes a motion velocity joints and the rotation angle of the body limb moving distance interval of temporally adjacent frames and the corresponding joint points calculated; the human joints and the body posture information of the standard motion model and motion parameters corresponding to a time point or time period compared acquires the difference.
  11. 11.如权利要求7所述的基于人体姿态识别的体育教学辅助方法,其特征在于,所述预定时间点为关键姿势的时间点,所述预定时间段为关键动作所处的时间段。 Physical teaching assistance method based on human gesture recognition, wherein said 7, the predetermined time point as the time point of key poses, the predetermined time period is a time period in which the operation key as claimed in claim.
  12. 12.如权利要求7所述的基于人体姿态识别的体育教学辅助方法,其特征在于,所述动作纠正提示为语音提示或图像提示。 Physical teaching assistance method based on human gesture recognition, wherein said 7, the operation to correct voice prompts to prompt or prompt image as claimed in claim.
CN 201210138503 2012-05-07 2012-05-07 Physical education assisting system and method based on human body posture recognition CN103390174A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 201210138503 CN103390174A (en) 2012-05-07 2012-05-07 Physical education assisting system and method based on human body posture recognition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 201210138503 CN103390174A (en) 2012-05-07 2012-05-07 Physical education assisting system and method based on human body posture recognition

Publications (1)

Publication Number Publication Date
CN103390174A true true CN103390174A (en) 2013-11-13

Family

ID=49534441

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 201210138503 CN103390174A (en) 2012-05-07 2012-05-07 Physical education assisting system and method based on human body posture recognition

Country Status (1)

Country Link
CN (1) CN103390174A (en)

Cited By (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103713545A (en) * 2013-12-17 2014-04-09 华为技术有限公司 Operation guiding method, device and system
CN104190063A (en) * 2014-08-08 2014-12-10 杭州电子科技大学 Posture recognition and remote training method and system based on smart phone
CN104318248A (en) * 2014-10-21 2015-01-28 北京智谷睿拓技术服务有限公司 Action recognition method and action recognition device
CN104307171A (en) * 2014-11-21 2015-01-28 京东方科技集团股份有限公司 Body sensing recognition device
CN104394386A (en) * 2014-12-11 2015-03-04 盐城工学院 System and method for evaluating knee joint motion normativity through multi-view sequence images
CN104392204A (en) * 2014-10-21 2015-03-04 北京智谷睿拓技术服务有限公司 Gesture recognition gesture recognition method and apparatus
CN104484574A (en) * 2014-12-25 2015-04-01 东华大学 Real-time human body gesture supervised training correction system based on quaternion
CN104474701A (en) * 2014-11-20 2015-04-01 杭州电子科技大学 Interactive system for promoting Taijiquan
CN104821116A (en) * 2015-05-04 2015-08-05 重庆科技学院 Prompting system of large-scale performance motions and prompting method thereof
CN104906776A (en) * 2014-03-13 2015-09-16 富泰华工业(深圳)有限公司 Somatosensory sports scoring system, device and method
CN104951081A (en) * 2015-07-09 2015-09-30 李乔亮 Method for automatically recognizing reading and writing postures and intelligent pre-warning device adopting method
CN105233465A (en) * 2015-11-11 2016-01-13 赫媛媛 Opera physique training device
CN105457254A (en) * 2016-01-14 2016-04-06 湖南师范大学 Intelligent training system for gymnastics training
CN105641899A (en) * 2014-12-31 2016-06-08 深圳泰山在线科技有限公司 Step physical fitness test method and system
CN105664462A (en) * 2016-01-07 2016-06-15 北京邮电大学 Auxiliary training system based on human body posture estimation algorithm
CN105788375A (en) * 2016-05-10 2016-07-20 山东交通学院 Multifunctional physical exercise teaching instrument
CN105791698A (en) * 2016-04-22 2016-07-20 周口职业技术学院 System and method of determining dance training
CN105847987A (en) * 2016-03-24 2016-08-10 乐视控股(北京)有限公司 Method and system for correcting human body actions through television and body feeling accessory component
CN106110627A (en) * 2016-06-20 2016-11-16 曲大方 Physical education and wushu sport movements correcting device and method
CN106139564A (en) * 2016-08-01 2016-11-23 纳恩博(北京)科技有限公司 Image processing method and device
CN106251721A (en) * 2016-09-26 2016-12-21 宇龙计算机通信科技(深圳)有限公司 Virtual motion teaching method and system
CN106464773A (en) * 2014-03-20 2017-02-22 2Mee有限公司 Augmented reality apparatus and method
CN106621281A (en) * 2016-12-21 2017-05-10 中国航天科工集团第四研究院指挥自动化技术研发与应用中心 Portable training evaluation system
CN106730770A (en) * 2017-01-03 2017-05-31 京东方科技集团股份有限公司 Motion posture detection apparatus, treadmill, and motion posture detection method

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101086681A (en) * 2006-06-09 2007-12-12 中国科学院自动化研究所 Game control system and method based on stereo vision
CN101202994A (en) * 2006-12-14 2008-06-18 北京三星通信技术研究有限公司;三星电子株式会社 Method and device assistant to user for body-building
KR20100131060A (en) * 2009-06-05 2010-12-15 삼성전자주식회사 Apparatus and method for video sensor-based human activity and facial expression modeling and recognition
CN101989326A (en) * 2009-07-31 2011-03-23 三星电子株式会社;北京三星通信技术研究有限公司 Human posture recognition method and device
CN102298649A (en) * 2011-10-09 2011-12-28 南京大学 A personal-space trajectory data retrieval operation method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101086681A (en) * 2006-06-09 2007-12-12 中国科学院自动化研究所 Game control system and method based on stereo vision
CN101202994A (en) * 2006-12-14 2008-06-18 北京三星通信技术研究有限公司;三星电子株式会社 Method and device assistant to user for body-building
KR20100131060A (en) * 2009-06-05 2010-12-15 삼성전자주식회사 Apparatus and method for video sensor-based human activity and facial expression modeling and recognition
CN101989326A (en) * 2009-07-31 2011-03-23 三星电子株式会社;北京三星通信技术研究有限公司 Human posture recognition method and device
CN102298649A (en) * 2011-10-09 2011-12-28 南京大学 A personal-space trajectory data retrieval operation method

Cited By (32)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103713545A (en) * 2013-12-17 2014-04-09 华为技术有限公司 Operation guiding method, device and system
CN104906776A (en) * 2014-03-13 2015-09-16 富泰华工业(深圳)有限公司 Somatosensory sports scoring system, device and method
CN104906776B (en) * 2014-03-13 2018-04-24 富泰华工业(深圳)有限公司 Somatosensory motion scoring system, apparatus and method
CN106464773A (en) * 2014-03-20 2017-02-22 2Mee有限公司 Augmented reality apparatus and method
CN104190063A (en) * 2014-08-08 2014-12-10 杭州电子科技大学 Posture recognition and remote training method and system based on smart phone
CN104190063B (en) * 2014-08-08 2016-07-06 杭州电子科技大学 Gesture recognition and remote training method and system based on smart phone
CN104392204B (en) * 2014-10-21 2018-03-09 北京智谷睿拓技术服务有限公司 Gesture recognition gesture recognition method and apparatus
CN104318248A (en) * 2014-10-21 2015-01-28 北京智谷睿拓技术服务有限公司 Action recognition method and action recognition device
CN104318248B (en) * 2014-10-21 2018-04-06 北京智谷睿拓技术服务有限公司 Motion recognition method and recognition apparatus operation
CN104392204A (en) * 2014-10-21 2015-03-04 北京智谷睿拓技术服务有限公司 Gesture recognition gesture recognition method and apparatus
CN104474701A (en) * 2014-11-20 2015-04-01 杭州电子科技大学 Interactive system for promoting Taijiquan
CN104307171B (en) * 2014-11-21 2017-08-08 京东方科技集团股份有限公司 Identification means somatosensory
CN104307171A (en) * 2014-11-21 2015-01-28 京东方科技集团股份有限公司 Body sensing recognition device
CN104394386A (en) * 2014-12-11 2015-03-04 盐城工学院 System and method for evaluating knee joint motion normativity through multi-view sequence images
CN104484574A (en) * 2014-12-25 2015-04-01 东华大学 Real-time human body gesture supervised training correction system based on quaternion
CN105641899B (en) * 2014-12-31 2017-11-24 深圳泰山体育科技股份有限公司 Physical testing step one kind of method and system
CN105641899A (en) * 2014-12-31 2016-06-08 深圳泰山在线科技有限公司 Step physical fitness test method and system
CN104821116A (en) * 2015-05-04 2015-08-05 重庆科技学院 Prompting system of large-scale performance motions and prompting method thereof
CN104951081B (en) * 2015-07-09 2016-04-13 李乔亮 Gesture recognition method for automatically reading and writing device and a smart warning
CN104951081A (en) * 2015-07-09 2015-09-30 李乔亮 Method for automatically recognizing reading and writing postures and intelligent pre-warning device adopting method
CN105233465B (en) * 2015-11-11 2018-02-16 赫媛媛 Opera physical training device
CN105233465A (en) * 2015-11-11 2016-01-13 赫媛媛 Opera physique training device
CN105664462A (en) * 2016-01-07 2016-06-15 北京邮电大学 Auxiliary training system based on human body posture estimation algorithm
CN105457254A (en) * 2016-01-14 2016-04-06 湖南师范大学 Intelligent training system for gymnastics training
CN105847987A (en) * 2016-03-24 2016-08-10 乐视控股(北京)有限公司 Method and system for correcting human body actions through television and body feeling accessory component
CN105791698A (en) * 2016-04-22 2016-07-20 周口职业技术学院 System and method of determining dance training
CN105788375A (en) * 2016-05-10 2016-07-20 山东交通学院 Multifunctional physical exercise teaching instrument
CN106110627A (en) * 2016-06-20 2016-11-16 曲大方 Physical education and wushu sport movements correcting device and method
CN106139564A (en) * 2016-08-01 2016-11-23 纳恩博(北京)科技有限公司 Image processing method and device
CN106251721A (en) * 2016-09-26 2016-12-21 宇龙计算机通信科技(深圳)有限公司 Virtual motion teaching method and system
CN106621281A (en) * 2016-12-21 2017-05-10 中国航天科工集团第四研究院指挥自动化技术研发与应用中心 Portable training evaluation system
CN106730770A (en) * 2017-01-03 2017-05-31 京东方科技集团股份有限公司 Motion posture detection apparatus, treadmill, and motion posture detection method

Similar Documents

Publication Publication Date Title
Fothergill et al. Instructing people for training gestural interactive systems
Zafrulla et al. American sign language recognition with the kinect
Chan et al. A virtual reality dance training system using motion capture technology
US6514081B1 (en) Method and apparatus for automating motion analysis
US20120092445A1 (en) Automatically tracking user movement in a video chat application
US20110292181A1 (en) Methods and systems using three-dimensional sensing for user interaction with applications
Riecke et al. Visual homing is possible without landmarks: A path integration study in virtual reality
US20100281432A1 (en) Show body position
US20130137076A1 (en) Head-mounted display based education and instruction
US20100167248A1 (en) Tracking and training system for medical procedures
US8418085B2 (en) Gesture coach
US20090258703A1 (en) Motion Assessment Using a Game Controller
US20130171596A1 (en) Augmented reality neurological evaluation method
US20120052947A1 (en) System and method for cyber training of martial art on network
KR20030085275A (en) Golf swing motion montoring method and system
US8953909B2 (en) System, method, and computer software code for mimic training
US7780450B2 (en) Video instructional system and method for teaching motor skills
Alexiadis et al. Evaluating a dancer's performance using kinect-based skeleton tracking
US6881067B2 (en) Video instructional system and method for teaching motor skills
Kwon et al. Combining body sensors and visual sensors for motion training
US8314840B1 (en) Motion analysis using smart model animations
Varona et al. Hands-free vision-based interface for computer accessibility
Motokawa et al. Support system for guitar playing using augmented reality display
US20130225305A1 (en) Expanded 3d space-based virtual sports simulation system
US20120108334A1 (en) Game device, control method for a game device, and a non-transitory information storage medium

Legal Events

Date Code Title Description
C06 Publication
C10 Entry into substantive examination
RJ01