CN107220608B - Basketball action model reconstruction and defense guidance system and method - Google Patents
Basketball action model reconstruction and defense guidance system and method Download PDFInfo
- Publication number
- CN107220608B CN107220608B CN201710362557.3A CN201710362557A CN107220608B CN 107220608 B CN107220608 B CN 107220608B CN 201710362557 A CN201710362557 A CN 201710362557A CN 107220608 B CN107220608 B CN 107220608B
- Authority
- CN
- China
- Prior art keywords
- basketball
- action
- movement
- joint points
- motions
- Prior art date
- Legal status (The legal status 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 status listed.)
- Expired - Fee Related
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/23—Recognition of whole body movements, e.g. for sport training
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B69/00—Training appliances or apparatus for special sports
- A63B69/0071—Training appliances or apparatus for special sports for basketball
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
- G06V20/42—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items of sport video content
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Physical Education & Sports Medicine (AREA)
- Computational Linguistics (AREA)
- Prostheses (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a basketball movement model reconstruction and defense guidance system, which comprises a Kinect depth sensor, a depth learning module and a basketball movement posture reconstruction module, wherein the Kinect depth sensor is used for extracting basketball movement information; the depth learning module classifies the basketball motions through depth learning, and stores depth values corresponding to skeleton coordinate points of the basketball motions in a series of pictures and coordinate values of a camera coordinate system in the Kinect depth sensor; the basketball movement posture reconstruction module is used for reconstructing a complete basketball movement according to the stored three-dimensional coordinates of the bone joint points; corresponding guidance is performed when the athlete trains. The invention combines the deep learning technology to realize the identification and statistics of basic basketball motions.
Description
Technical Field
The invention relates to the field of deep learning detection, in particular to a basketball action model reconstruction and defense guidance system and method.
Background
At present, the development of the Kinect sensor technology promotes the model reconstruction technology, but the prior art only detects the action of a person and is not applied to the basketball field, so that a real person can perform targeted defense on the reconstructed model.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a basketball movement model reconstruction and defense guidance system, which can obtain more accurate basketball movement information, thereby reconstructing a model in space and carrying out targeted training for athletes.
The invention also aims to provide a basketball motion model reconstruction and defense guiding method.
The purpose of the invention is realized by the following technical scheme:
a basketball motion model reconstruction and defense guidance system comprises a Kinect depth sensor, a deep learning module and a basketball motion posture reconstruction module; wherein
The Kinect depth sensor is used for extracting the complete basketball movement color image, the basketball movement bone characteristics and the depth information of the basketball movement bone joint points of the athlete;
the deep learning module is used for training the correct basketball movement through deep learning to obtain the weight so as to classify the basketball movement; then, recognizing the input basketball action color map, and if the required basketball action is recognized, extracting the basketball action bone characteristics and the depth information of the basketball action bone joint points corresponding to the basketball action through a Kinect depth sensor; thus, a complete basketball action picture set of continuous frames is obtained, and the depth values corresponding to the basketball action skeleton coordinate points in the series of pictures and the coordinate values of the camera coordinate system in the Kinect depth sensor are stored;
the basketball movement posture reconstruction module is used for reconstructing a complete basketball movement according to the stored three-dimensional coordinates of the bone joint points; when a player trains corresponding basketball motions, a model of the corresponding basketball motions is simulated, the main skeleton joint points of the player are trained at the same depth, and whether the corresponding basketball motions are standard or not is predicted according to the extraction of the three-dimensional coordinates of the skeleton joint points of the player.
Most people have specific muscle memory during sports, so most athletes have the same action amplitude during shooting and breaking through, and a defender trains defending time according to a model to carry out the defending memory required by capping and breaking, thereby providing help for actual competition.
The deep learning refers to a deep learning method based on the yolo framework. The positions of the players can be accurately found and the basketball motions can be distinguished.
The basketball action comprises cap covering, shooting and snapping.
The other purpose of the invention is realized by the following technical scheme:
a basketball movement model reconstruction and defense guiding method comprises the following steps:
s1, extracting the complete basketball movement color map, the basketball movement skeleton characteristics and the depth information of the basketball movement skeleton joint points of the athlete;
s2, training the correct basketball movement through deep learning, and then obtaining weights so as to classify the basketball movement; then, recognizing the input basketball action color map, and if the required basketball action is recognized, extracting the basketball action bone characteristics and the depth information of the basketball action bone joint points corresponding to the basketball action through a Kinect depth sensor; thus, a complete basketball action picture set of continuous frames is obtained, and the depth values corresponding to the basketball action skeleton coordinate points in the series of pictures and the coordinate values of the camera coordinate system in the Kinect depth sensor are stored;
s3, reconstructing a complete basketball movement according to the stored three-dimensional coordinates of the bone joint points; when a player trains corresponding basketball motions, a model of the corresponding basketball motions is simulated, the main skeleton joint points of the player are trained at the same depth, and whether the corresponding basketball motions are standard or not is predicted according to the extraction of the three-dimensional coordinates of the skeleton joint points of the player.
Compared with the prior art, the invention has the following advantages and beneficial effects:
(1) the method uses the Kinect sensor technology, and only three-dimensional coordinates are needed, so that accurate model reconstruction of basketball movement needing to be researched is realized.
(2) The invention can lead the athlete to have more targeted defense training.
(3) The invention does not need to install any body sensing sensor on the body of the observer.
(4) The invention improves the efficiency of identifying the color picture through the deep learning detection technology.
Drawings
Fig. 1 is a flowchart of a basketball movement model reconstruction and defense guidance method according to the present invention.
Fig. 2 shows depth information of basketball action bone joint points extracted under the Kinect depth sensor.
FIG. 3 is a basketball action bone feature map extracted under the Kinect depth sensor.
Detailed Description
The present invention will be described in further detail with reference to examples and drawings, but the present invention is not limited thereto.
As shown in fig. 1, in order to reconstruct the posture of the basketball movement, the color map information, the depth map information and the skeletal features of the basketball movement are obtained by the Kinect depth sensor. Gather extensive standard basketball action, based on yolo frame, train the weight that needs to classify the basketball action, instruct the defensive person to carry out special training at last.
As shown in fig. 2, the subject is performing a shooting operation, and continuous several frames of pictures that are considered to be highly likely to perform the shooting operation are extracted by the detection of deep learning, and the three-dimensional coordinates of the skeletal joint points are acquired. Thus, a basic model can be obtained. When a defender carries out cap training under the Kinect depth sensor, a model for virtually simulating a series of basketball motions of the basketball shooting person can be obtained, the main skeleton joint points of the trainer are trained under the same depth, and whether defense succeeds or not is predicted according to the extraction of the three-dimensional coordinates of the skeleton joint points of the defender. Most people have specific muscle memory during sports, so most athletes have the same action amplitude during shooting and breaking through, and a defender trains defending time according to a model to carry out the defending memory required by capping and breaking, thereby providing help for actual competition.
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such changes, modifications, substitutions, combinations, and simplifications are intended to be included in the scope of the present invention.
Claims (3)
1. A basketball movement model reconstruction and defense guidance system is characterized in that: the basketball movement posture rebuilding system comprises a Kinect depth sensor, a deep learning module and a basketball movement posture rebuilding module; wherein
The Kinect depth sensor is used for extracting the complete basketball movement color image, the basketball movement bone characteristics and the depth information of the basketball movement bone joint points of the athlete;
the deep learning module is used for training the correct basketball movement through deep learning to obtain the weight so as to classify the basketball movement; then, recognizing the input basketball action color map, and if the required basketball action is recognized, extracting the basketball action bone characteristics and the depth information of the basketball action bone joint points corresponding to the basketball action through a Kinect depth sensor; thus, a complete basketball action picture set of continuous frames is obtained, and the depth values corresponding to the basketball action skeleton coordinate points in the series of pictures and the coordinate values of the camera coordinate system in the Kinect depth sensor are stored; the basketball movement posture reconstruction module is used for reconstructing a complete basketball movement according to the stored three-dimensional coordinates of the bone joint points; when a player trains corresponding basketball motions, a model of the corresponding basketball motions is simulated, main skeleton joint points of the player are trained at the same depth, and whether the corresponding basketball motions are standard or not is predicted according to extraction of three-dimensional coordinates of the skeleton joint points of the player; when a defender carries out cap training under a Kinect depth sensor, a model for a series of basketball motions of the shooter is simulated virtually, main skeleton joint points of the trainer are trained under the same depth, and whether the defending is successful or not is predicted according to the extraction of three-dimensional coordinates of the skeleton joint points of the defender;
the deep learning refers to a deep learning method based on the yolo framework.
2. The basketball motion model reconstruction and defense guidance system as claimed in claim 1, further comprising: the basketball action comprises cap covering, shooting and snapping.
3. A method for guiding the reconstruction and defense of a basketball movement model based on the system for guiding the reconstruction and defense of a basketball movement model according to any one of claims 1 to 2, comprising the steps of:
s1, extracting the complete basketball movement color map, the basketball movement skeleton characteristics and the depth information of the basketball movement skeleton joint points of the athlete;
s2, training the correct basketball movement through deep learning, and then obtaining weights so as to classify the basketball movement; then, recognizing the input basketball action color map, and if the required basketball action is recognized, extracting the basketball action bone characteristics and the depth information of the basketball action bone joint points corresponding to the basketball action through a Kinect depth sensor; thus, a complete basketball action picture set of continuous frames is obtained, and the depth values corresponding to the basketball action skeleton coordinate points in the series of pictures and the coordinate values of the camera coordinate system in the Kinect depth sensor are stored;
s3, reconstructing a complete basketball movement according to the stored three-dimensional coordinates of the bone joint points; when a player trains corresponding basketball motions, a model of the corresponding basketball motions is simulated, the main skeleton joint points of the player are trained at the same depth, and whether the corresponding basketball motions are standard or not is predicted according to the extraction of the three-dimensional coordinates of the skeleton joint points of the player.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710362557.3A CN107220608B (en) | 2017-05-22 | 2017-05-22 | Basketball action model reconstruction and defense guidance system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710362557.3A CN107220608B (en) | 2017-05-22 | 2017-05-22 | Basketball action model reconstruction and defense guidance system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107220608A CN107220608A (en) | 2017-09-29 |
| CN107220608B true CN107220608B (en) | 2021-06-08 |
Family
ID=59944835
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710362557.3A Expired - Fee Related CN107220608B (en) | 2017-05-22 | 2017-05-22 | Basketball action model reconstruction and defense guidance system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107220608B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107909060A (en) * | 2017-12-05 | 2018-04-13 | 前海健匠智能科技（深圳）有限公司 | Gymnasium body-building action identification method and device based on deep learning |
| CN109821243A (en) * | 2019-01-25 | 2019-05-31 | 丰羽教育科技（上海）有限公司 | A method of simulation reappears shooting process |
| CN109948459B (en) * | 2019-02-25 | 2023-08-25 | 广东工业大学 | Football action evaluation method and system based on deep learning |
| CN110478883B (en) * | 2019-08-21 | 2021-04-13 | 南京信息工程大学 | Body-building action teaching and correcting system and method |
| CN110960843A (en) * | 2019-12-23 | 2020-04-07 | 天水师范学院 | Basketball skill auxiliary training system |
| CN111569398B (en) * | 2020-05-09 | 2022-01-04 | 深圳市洲明科技股份有限公司 | Semi-immersion type bowling training system and method based on LED display screen |
| CN117475514A (en) * | 2023-11-10 | 2024-01-30 | 广州市微锋科技有限公司 | Shooting training system and method based on image analysis |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104850846A (en) * | 2015-06-02 | 2015-08-19 | 深圳大学 | Human behavior recognition method and human behavior recognition system based on depth neural network |
| CN104899561A (en) * | 2015-05-27 | 2015-09-09 | 华南理工大学 | Parallelized human body behavior identification method |
| CN106650562A (en) * | 2016-06-14 | 2017-05-10 | 西安电子科技大学 | Online continuous human behavior identification method based on Kinect |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8408982B2 (en) * | 2007-05-24 | 2013-04-02 | Pillar Vision, Inc. | Method and apparatus for video game simulations using motion capture |
| US9639746B2 (en) * | 2011-07-28 | 2017-05-02 | Arb Labs Inc. | Systems and methods of detecting body movements using globally generated multi-dimensional gesture data |
| US9625994B2 (en) * | 2012-10-01 | 2017-04-18 | Microsoft Technology Licensing, Llc | Multi-camera depth imaging |
| WO2014194337A1 (en) * | 2013-05-30 | 2014-12-04 | Atlas Wearables, Inc. | Portable computing device and analyses of personal data captured therefrom |
| US9734405B2 (en) * | 2015-10-05 | 2017-08-15 | Pillar Vision, Inc. | Systems and methods for monitoring objects in athletic playing spaces |
| CN105512621B (en) * | 2015-11-30 | 2019-04-09 | 华南理工大学 | A kind of shuttlecock action director's system based on Kinect |
| CN106022213B (en) * | 2016-05-04 | 2019-06-07 | 北方工业大学 | A kind of human motion recognition method based on three-dimensional bone information |
| CN105999670B (en) * | 2016-05-31 | 2018-09-07 | 山东科技大学 | Taijiquan action based on kinect judges and instructs system and its guidance method |
| CN106581949B (en) * | 2016-11-24 | 2019-04-30 | 石家庄铁路职业技术学院 | A kind of shootaround monitoring and evaluating system |
| CN106650687B (en) * | 2016-12-30 | 2020-05-19 | 山东大学 | Posture correction method based on depth information and skeleton information |
-
2017
- 2017-05-22 CN CN201710362557.3A patent/CN107220608B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104899561A (en) * | 2015-05-27 | 2015-09-09 | 华南理工大学 | Parallelized human body behavior identification method |
| CN104850846A (en) * | 2015-06-02 | 2015-08-19 | 深圳大学 | Human behavior recognition method and human behavior recognition system based on depth neural network |
| CN106650562A (en) * | 2016-06-14 | 2017-05-10 | 西安电子科技大学 | Online continuous human behavior identification method based on Kinect |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107220608A (en) | 2017-09-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107220608B (en) | Basketball action model reconstruction and defense guidance system and method | |
| CN109191588B (en) | Motion teaching method, motion teaching device, storage medium and electronic equipment | |
| WO2021051579A1 (en) | Body pose recognition method, system, and apparatus, and storage medium | |
| KR102106135B1 (en) | Apparatus and method for providing application service by using action recognition | |
| US20180357472A1 (en) | Systems and methods for creating target motion, capturing motion, analyzing motion, and improving motion | |
| US11798318B2 (en) | Detection of kinetic events and mechanical variables from uncalibrated video | |
| CN106166376B (en) | Simplify taijiquan in 24 forms comprehensive training system | |
| CN110448870B (en) | Human body posture training method | |
| CN110544301A (en) | Three-dimensional human body action reconstruction system, method and action training system | |
| CN109214231A (en) | Physical education auxiliary system and method based on human body attitude identification | |
| CN108096807A (en) | A kind of exercise data monitoring method and system | |
| CN110427900B (en) | Method, device and equipment for intelligently guiding fitness | |
| CN103354761A (en) | Virtual golf simulation apparatus and method | |
| CN104035557A (en) | Kinect action identification method based on joint activeness | |
| KR100907704B1 (en) | Golfer's posture correction system using artificial caddy and golfer's posture correction method using it | |
| CN110782482A (en) | Motion evaluation method and device, computer equipment and storage medium | |
| US20220366653A1 (en) | Full Body Virtual Reality Utilizing Computer Vision From a Single Camera and Associated Systems and Methods | |
| CN112007343A (en) | Double-arm boxing training robot | |
| CN112933581A (en) | Sports action scoring method and device based on virtual reality technology | |
| CN116328279A (en) | Real-time auxiliary training method and device based on visual human body posture estimation | |
| CN103310191A (en) | Human body action identification method for motion information imaging | |
| CN109407826B (en) | Ball game simulation method and device, storage medium and electronic equipment | |
| Yang et al. | Research on face recognition sports intelligence training platform based on artificial intelligence | |
| CN103218826B (en) | Projectile based on Kinect detection, three-dimensional localization and trajectory predictions method | |
| CN116271757A (en) | Auxiliary system and method for basketball practice based on AI technology |
Legal Events
| 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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210608 |