CN108021883A - The method, apparatus and storage medium of sphere recognizing model of movement - Google Patents
The method, apparatus and storage medium of sphere recognizing model of movement Download PDFInfo
- Publication number
- CN108021883A CN108021883A CN201711259464.4A CN201711259464A CN108021883A CN 108021883 A CN108021883 A CN 108021883A CN 201711259464 A CN201711259464 A CN 201711259464A CN 108021883 A CN108021883 A CN 108021883A
- Authority
- CN
- China
- Prior art keywords
- court
- sphere
- convolution kernel
- displacement
- model
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of method, apparatus and storage medium of sphere recognizing model of movement, this method includes：Using local multiframe trace information, the position of best match and displacement model are searched for by defining different positions and displacement convolution kernel；According to the position of best match and displacement model, judgement obtains stable sphere motor pattern.The method of the present invention can remove the noise jamming in track, stablize the motion state for identifying sphere by local multiframe trace information；The method of the present invention need not establish positive negative example base and carry out training pattern, easy to operate.And this method has wide range of applications, suitable for the identification of most sphere motor pattern.
Description
Technical field
The present invention relates to intelligent sphere field technology field, more particularly to a kind of method, apparatus of sphere recognizing model of movement and
Storage medium.
Background technology
The identification of sphere motor pattern is a part indispensable in intelligent court, is the base of sphere gripper path analysis
Plinth, tool play a very important role.Identification generally use the following two kinds scheme to sphere motor pattern at present：
First, by establishing sample storehouse, Ma Erliefu models are implied in training, identify the motor pattern of sphere；Second, by building
Vertical sample storehouse, trains corresponding neural network classifier, identifies the motor pattern of sphere.
But existing method such as implicit Ma Erliefu models and neural network classifier be required for establishing positive negative example base into
Row training, thus flow is more complicated.In addition, existing part sphere trajectory analysis method using two continuous frames trace information into
The identification of row sphere motor pattern, easily by noise jamming, causes to identify unstable.
The content of the invention
The present invention provides a kind of method, apparatus for the sphere recognizing model of movement that flow is simple, identification is stable and storage is situated between
Matter.
To achieve the above object, the present invention provides a kind of method of sphere recognizing model of movement, including：
Using local multiframe trace information, the position of best match is searched for by defining different positions and displacement convolution kernel
Put and displacement model；
According to the position of best match and displacement model, judgement obtains stable sphere motor pattern.
Wherein, it is described using local multiframe trace information, searched for most by defining different positions and displacement convolution kernel
The step of good matched position and displacement model, includes：
The three-dimensional track of sphere is extracted by three-dimensional reconstruction；
Based on the three-dimensional track of the sphere extracted, local multiframe trace information is obtained；
The different position of the sphere motor pattern and displacement convolution kernel are defined, different spheres are calculated by convolution kernel
The matching degree of motor pattern；
Using the local multiframe trace information, calculated by the different position of the sphere of definition and displacement convolution kernel
Corresponding matching degree is obtained, then searches for position and the displacement model of best match.
Wherein, the displacement convolution kernel for defining the sphere motor pattern includes：The convolution defined based on location status
Core, the convolution kernel defined based on displacement state.
Wherein, it is described to be included based on the convolution kernel that location status defines：A half-court convolution kernel, B half-court convolution kernel, A half-court mistakes
Net crosses net to A half-court convolution kernels to B half-court convolution kernel, B half-court.
Wherein, it is described to be included based on the convolution kernel that displacement state defines：A half-court moves to B half-court convolution kernel, B half-court fortune
Move A half-court convolution kernel, A to B and then B to A convolution kernels, B to A and then A to B convolution kernels.
Wherein, the sphere motor pattern includes two types：Action and stable state, wherein：
Action includes：A half-court bounce the ball action, A half-court to B half-court of action, B half-court of bouncing the ball crosses net action, B half-court to A
Half-court crosses net action, the action that A half-court is touched net, the action of B half-court net-fault；
Stable state includes：Tennis is invisible, tennis moves to B half-court, tennis in A half-court simultaneously in A half-court and from A half-court
A half-court, tennis are moved in B half-court from B half-court and are moved to A half-court, tennis in B half-court from B half-court and moved to from A half-court
B half-court.
In addition, the present invention also proposes a kind of device of sphere recognizing model of movement, including memory, processor and storage
Computer program on the memory, the computer program realize method as described above when being run by the processor
The step of.
In addition, the present invention also proposes a kind of computer-readable recording medium, stored on the computer-readable recording medium
There is the step of computer program, the computer program realizes method as described above when being run by processor.
Compared with prior art, existing method such as implicit Ma Erliefu models and neural network classifier is required for establishing positive and negative
Sample storehouse is trained, thus flow is more complicated, and the method for the present invention need not establish positive negative example base and carry out training pattern, and move
Pattern-recognition is more stable, has the characteristics that simple and stable, achievees the purpose that easy to operate and general.Existing part sphere rail
Mark analysis method carries out the identification of sphere motor pattern using the trace information of two continuous frames, easily by noise jamming, causes to know
Not unstable, the method for the present invention achievees the purpose that moving state identification is more stable using the trace information of local multiframe.
Therefore, local multiframe rail is passed through compared to the method being identified by two continuous frames trace information, the method for the present invention
Mark information, can remove the noise jamming in track, stablize the motion state for identifying sphere.And this method has a wide range of application
It is general, suitable for the identification of most sphere motor pattern.
Brief description of the drawings
Fig. 1 is the flow diagram of the embodiment of the method for sphere recognizing model of movement proposed by the present invention；
Fig. 2 is position view of the sphere recognizing model of movement of the present invention in trajectory analysis；
Fig. 3 is sphere recognizing model of movement flow chart.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
Specifically, Fig. 1 is refer to, Fig. 1 is the flow of the embodiment of the method for sphere recognizing model of movement proposed by the present invention
Schematic diagram.
As shown in Figure 1, the embodiment of the present invention proposes a kind of method of sphere recognizing model of movement, including：
Step S1, using local multiframe trace information, is searched for optimal by defining different positions and displacement convolution kernel
Matched position and displacement model；
Specifically, the three-dimensional track of sphere is extracted by three-dimensional reconstruction；Three-dimensional rail based on the sphere extracted
Mark, obtains local multiframe trace information；The different position of the sphere motor pattern and displacement convolution kernel are defined, passes through convolution kernel
The matching degree of sphere motor pattern is calculated；It is different by the sphere of definition using the local multiframe trace information
Position and displacement convolution kernel corresponding matching degree is calculated, then search for best match position and displacement model.
Wherein, the displacement convolution kernel for defining the sphere motor pattern includes：The convolution defined based on location status
Core, the convolution kernel defined based on displacement state.
It is described to be included based on the convolution kernel that location status defines：A half-court convolution kernel, B half-court convolution kernel, A half-court cross net to B
Half-court convolution kernel, B half-court cross net to A half-court convolution kernels.
It is described to be included based on the convolution kernel that displacement state defines：A half-court moves to B half-court convolution kernel, B half-court moves to A
Half-court convolution kernel, A to B and then B to A convolution kernels, B to A and then A to B convolution kernels.
Step S2, according to the position of best match and displacement model, judgement obtains stable sphere motor pattern.
Usual sphere motor pattern includes two types：Action and stable state, wherein：
Action includes：A half-court bounce the ball action, A half-court to B half-court of action, B half-court of bouncing the ball crosses net action, B half-court to A
Half-court crosses net action, the action that A half-court is touched net, the action of B half-court net-fault；
Stable state includes：Tennis is invisible, tennis moves to B half-court, tennis in A half-court simultaneously in A half-court and from A half-court
A half-court, tennis are moved in B half-court from B half-court and are moved to A half-court, tennis in B half-court from B half-court and moved to from A half-court
B half-court.
According to the position of best match and displacement model, sphere motor pattern is identified.
Compared with prior art, existing method such as implicit Ma Erliefu models and neural network classifier is required for establishing positive and negative
Sample storehouse is trained, thus flow is more complicated, and the method for the present invention need not establish positive negative example base and carry out training pattern, and move
Pattern-recognition is more stable, has the characteristics that simple and stable, achievees the purpose that easy to operate and general.Existing part sphere rail
The identification of sphere motor pattern is carried out using the trace information of two continuous frames in mark analysis method, easily by noise jamming, is caused
Identify unstable, the method for the present invention achievees the purpose that moving state identification is more stable using the trace information of local multiframe.
Therefore, local multiframe rail is passed through compared to the method being identified by two continuous frames trace information, the method for the present invention
Mark information, can remove the noise jamming in track, stablize the motion state for identifying sphere.And this method has a wide range of application
It is general, suitable for the identification of most sphere motor pattern.
The concrete scheme of identification sphere motor pattern of the embodiment of the present invention is described in detail below：
Wherein, the basis of sphere recognizing model of movement is：The three-dimensional track of sphere has been extracted by three-dimensional reconstruction；Sphere
The basis that recognizing model of movement will judge and score as next part out-of-bounds, refers to Fig. 2.
Part existing method uses the trace information of two continuous frames, but identification is unstable, easily by noise jamming.
Therefore this method uses the trace information and displacement information of local multiframe.In order to make the more generalization of the flow of recognizing model of movement,
Convolution kernel is defined according to motor pattern, the matching degree of the motor pattern is then calculated by convolution kernel, searches for optimal
With pattern, the motor pattern of sphere is obtained according to the best match pattern-recognition of position and displacement.
The present invention includes three parts：The definition of convolution kernel, the identification of best match pattern search, sphere motor pattern,
Refer to Fig. 3.
1st, the convolution kernel definition of motor pattern
The convolution kernel defined based on location status：
(1) A half-court convolution kernel defines：{-1,-1,-1,-1,-1,-1,-1,-1,-1}
(2) B half-court convolution kernel defines：{1,1,1,1,1,1,1,1,1}
(3) A half-court is crossed net and is defined to B half-court convolution kernel：{1,1,1,1,1,-1,-1,-1,-1}
(4) B half-court is crossed net and is defined to A half-court convolution kernel：{-1,-1,-1,-1,-1,1,1,1,1}
The corresponding four kinds of mode positions (Pos_Model_Type) of these four convolution kernels, respectively with formula POS_ASIDE, POS_
BSIDE, POS_NETAB, POS_NETBA are represented.
The convolution kernel defined based on displacement state：
(1) A half-court moves to the definition of B half-court convolution kernel：{1,1,1,1,1,1,1,1}
(2) B half-court moves to the definition of A half-court convolution kernel：{-1,-1,-1,-1,-1,-1,-1,-1}
(3) A to B and then the definition of B to A convolution kernels：{-1,-1,-1,-1,1,1,1,1}
(4) B to A and then the definition of A to B convolution kernels：{1,1,1,1,-1,-1,-1,-1}
The corresponding four kinds of displacement models (Direct_Model_Type) of these four convolution kernels, respectively with DIRECT_ATOB,
DIRECT_BTOA, DIRECT_ABTOBA, DIRECT_BATOAB are represented.
2 best match pattern searches
2.1 location-based best match pattern searches
Local N frame positions：{(x1,y1, z1),…,(xN,yN,zN)}；
(1) convolution kernel PosKernel [j, i], wherein j are the type of position convolution kernel, and i is i-th of value of the convolution kernel；
(2)MatchjFor the matching degree of jth kind pattern, location-based pattern match degree calculates：
(3) in M kind mode positions POS_ASIDE, POS_BSIDE, POS_NETAB, POS_NETBA, optimal is searched for
The pattern matched somebody with somebody：
MatchBest=Max (Match1,…,MatchM)
Then the best match pattern using the pattern Pos_Model_Type corresponding to MatchBest as position.
The 2.2 best match pattern searches based on displacement
(1) displacement of part N frames：{(dx1,dy1,dz1),…,(dxN-1,dyN-1,dzN-1), wherein：
(2) convolution kernel DirectKernel [j, i], wherein j are the type of displacement convolution kernel, and i is the convolution kernel i-th
Put.
(3)MatchjFor the matching degree of jth kind pattern, the pattern match degree based on displacement calculates：
(4) in M kind mode positions DIRECT_ATOB, DIRECT_BTOA, DIRECT_ABTOBA, DIRECT_BATOAB
In, search for the pattern of best match：
MatchBest=Max (Match1,…,MatchM)；
Then the best match pattern using the pattern Direct_Model_Type corresponding to MatchBest as displacement.
The recognizing model of movement of 3 spheres
The motor pattern of sphere includes two types：Action and stable state.
Action includes：A half-court bounce the ball action, A half-court to B half-court of action, B half-court of bouncing the ball crosses net action, B half-court to A
Half-court crosses net action, the action that A half-court is touched net, the action of B half-court net-fault.
Stable state includes：Tennis is invisible, tennis moves to B half-court, tennis in A half-court simultaneously in A half-court and from A half-court
A half-court, tennis are moved in B half-court from B half-court and are moved to A half-court, tennis in B half-court from B half-court and moved to from A half-court
B half-court.
The motor pattern of sphere by the best match pattern recognition result of position and displacement, will be identified below.
The definition of 3.1 sphere motor patterns
All sphere motor patterns：
enumemMotionModelType
{
NOT_VISBLIE=-1, // ball are invisible
A_ATOB=0, // ball move to B half-court in A half-court and from A half-court
A_BTOA=1, // ball move to A half-court in A half-court and from B half-court
B_BTOA=2, // ball move to A half-court in B half-court and from B half-court
B_ATOB=3//ball moves to B half-court in B half-court and from A half-court
A_HIT=4, //A half-court are bounced the ball action
B_HIT=5, //B half-court are bounced the ball action
OVER_NET_AB=6, //A half-court to B half-court cross net action
OVER_NET_BA=7, //B half-court to A half-court cross net action
A_HITNET=8, the action that //A half-court is touched net
The action that B_HITNET=9//B half-court is touched net
}；
3.2 recognizing model of movement
The part needs to judge not by Pos_Model_Type mode positions and Direct_Model_Type displacement models
Same motor pattern：
This method can remove the noise jamming in track, stabilization identifies sphere by local multiframe trace information
Motion state.And this method has wide range of applications, suitable for the identification of most sphere motor pattern.It is implicit compared to existing
The recognition methods of Ma Erliefu models and neural network classifier, this method have the characteristics of easy operation.
In addition, the present invention also proposes a kind of device of sphere recognizing model of movement, including memory, processor and storage
Computer program on the memory, realizes when the computer program is run by the processor and is performed as described above described in example
Method the step of.
The present invention also proposes a kind of computer-readable recording medium, and calculating is stored with the computer-readable recording medium
Machine program, realizes the step of method described in example is performed as described above when the computer program is run by processor.
Compared to the method being identified by two continuous frames trace information, this method passes through local multiframe trace information, energy
The noise jamming in track is enough removed, stablizes the motion state for identifying sphere.And this method has wide range of applications, it is suitable for
The identification of most sphere motor pattern.Compared to the identification side of existing implicit Ma Erliefu models and neural network classifier
Method, this method have the characteristics of easy operation, it is not necessary to establish positive negative example base and carry out training pattern, and recognizing model of movement compares
Stablize.
The foregoing is merely the preferred embodiment of the present invention, is not intended to limit the scope of the invention, every utilization
Equivalent structure or the flow conversion that description of the invention and accompanying drawing content are made, are directly or indirectly used in other relevant skills
Art field, is included within the scope of the present invention.
Claims (8)
- A kind of 1. method of sphere recognizing model of movement, it is characterised in that including：Using local multiframe trace information, by define different positions and displacement convolution kernel search for the position of best match and Displacement model；According to the position of best match and displacement model, judgement obtains stable sphere motor pattern.
- 2. the method for sphere recognizing model of movement according to claim 1, it is characterised in that described to use local multiframe rail Mark information, is wrapped by defining different positions and displacement convolution kernel to search for the step of the position of best match and displacement model Include：The three-dimensional track of sphere is extracted by three-dimensional reconstruction；Based on the three-dimensional track of the sphere extracted, local multiframe trace information is obtained；The different position of the sphere motor pattern and displacement convolution kernel are defined, different spheres, which are calculated, by convolution kernel moves The matching degree of pattern；Using the local multiframe trace information, it is calculated by the different position of the sphere of definition and displacement convolution kernel Corresponding matching degree, then searches for position and the displacement model of best match.
- 3. the method for sphere recognizing model of movement according to claim 2, it is characterised in that described to define the sphere fortune The displacement convolution kernel of dynamic model formula includes：The convolution kernel defined based on location status, the convolution kernel defined based on displacement state.
- 4. the method for sphere recognizing model of movement according to claim 3, it is characterised in that described to be determined based on location status The convolution kernel of justice includes：A half-court convolution kernel, B half-court convolution kernel, A half-court cross net and cross net to A to B half-court convolution kernel and B half-court Half-court convolution kernel.
- 5. the method for sphere recognizing model of movement according to claim 4, it is characterised in that described to be determined based on displacement state The convolution kernel of justice includes：A half-court moves to B half-court convolution kernel, B half-court moves to A half-court convolution kernel, A to B and then B to A convolution Core, B to A and then A to B convolution kernels.
- 6. the method for sphere recognizing model of movement according to claim 4, it is characterised in that the sphere motor pattern bag Containing two types：Action and stable state, wherein：Action includes：A half-court bounce the ball action, A half-court to B half-court of action, B half-court of bouncing the ball crosses net action, B half-court to A half-court Cross net action, A half-court touch net action, B half-court touch net action；Stable state includes：Tennis is invisible, tennis moves to B half-court, tennis in A half-court and from B in A half-court and from A half-court Half-court moves to A half-court, tennis in B half-court and moves to A half-court, tennis in B half-court from B half-court and move to B half from A half-court .
- 7. a kind of device of sphere recognizing model of movement, it is characterised in that including memory, processor and be stored in described deposit Computer program on reservoir, realizes such as claim 1-6 any one when the computer program is run by the processor The step of described method.
- 8. a kind of computer-readable recording medium, it is characterised in that be stored with computer on the computer-readable recording medium Program, the step of method as any one of claim 1-6 is realized when the computer program is run by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711259464.4A CN108021883B (en) | 2017-12-04 | 2017-12-04 | Method, device and storage medium for recognizing movement pattern of sphere |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711259464.4A CN108021883B (en) | 2017-12-04 | 2017-12-04 | Method, device and storage medium for recognizing movement pattern of sphere |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108021883A true CN108021883A (en) | 2018-05-11 |
| CN108021883B CN108021883B (en) | 2020-07-21 |
Family
ID=62078177
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711259464.4A Active CN108021883B (en) | 2017-12-04 | 2017-12-04 | Method, device and storage medium for recognizing movement pattern of sphere |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108021883B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101354786A (en) * | 2007-07-23 | 2009-01-28 | 中国科学院计算技术研究所 | Analysis method of sports video case |
| CN103440277A (en) * | 2013-08-12 | 2013-12-11 | 合肥寰景信息技术有限公司 | Action model feature library and construction method thereof |
| CN104966045A (en) * | 2015-04-02 | 2015-10-07 | 北京天睿空间科技有限公司 | Video-based airplane entry-departure parking lot automatic detection method |
| CN105118343A (en) * | 2015-09-15 | 2015-12-02 | 北京瑞盖科技有限公司 | Tennis training examination and evaluation system and training method |
| CN105405150A (en) * | 2015-10-21 | 2016-03-16 | 东方网力科技股份有限公司 | Abnormal behavior detection method and abnormal behavior detection device based fused characteristics |
| CN106131469A (en) * | 2016-06-24 | 2016-11-16 | 北京天天乐动科技有限公司 | Ball intelligent robot based on machine vision coach and judgment system |
| CN106780620A (en) * | 2016-11-28 | 2017-05-31 | 长安大学 | A kind of table tennis track identification positioning and tracking system and method |
| CN107274433A (en) * | 2017-06-21 | 2017-10-20 | 吉林大学 | Method for tracking target, device and storage medium based on deep learning |
| CN107403167A (en) * | 2017-08-03 | 2017-11-28 | 华中师范大学 | Gesture identification method and device |
-
2017
- 2017-12-04 CN CN201711259464.4A patent/CN108021883B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101354786A (en) * | 2007-07-23 | 2009-01-28 | 中国科学院计算技术研究所 | Analysis method of sports video case |
| CN103440277A (en) * | 2013-08-12 | 2013-12-11 | 合肥寰景信息技术有限公司 | Action model feature library and construction method thereof |
| CN104966045A (en) * | 2015-04-02 | 2015-10-07 | 北京天睿空间科技有限公司 | Video-based airplane entry-departure parking lot automatic detection method |
| CN105118343A (en) * | 2015-09-15 | 2015-12-02 | 北京瑞盖科技有限公司 | Tennis training examination and evaluation system and training method |
| CN105405150A (en) * | 2015-10-21 | 2016-03-16 | 东方网力科技股份有限公司 | Abnormal behavior detection method and abnormal behavior detection device based fused characteristics |
| CN106131469A (en) * | 2016-06-24 | 2016-11-16 | 北京天天乐动科技有限公司 | Ball intelligent robot based on machine vision coach and judgment system |
| CN106780620A (en) * | 2016-11-28 | 2017-05-31 | 长安大学 | A kind of table tennis track identification positioning and tracking system and method |
| CN107274433A (en) * | 2017-06-21 | 2017-10-20 | 吉林大学 | Method for tracking target, device and storage medium based on deep learning |
| CN107403167A (en) * | 2017-08-03 | 2017-11-28 | 华中师范大学 | Gesture identification method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108021883B (en) | 2020-07-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kitsikidis et al. | Dance analysis using multiple kinect sensors | |
| CN103246884B (en) | Real-time body's action identification method based on range image sequence and device | |
| US8634638B2 (en) | Real-time action detection and classification | |
| CN103325125B (en) | Based on the motion target tracking method improving multi-instance learning algorithm | |
| CN102737386B (en) | Tracking is blocked in a kind of anti-fusion of moving target | |
| CN107103326A (en) | The collaboration conspicuousness detection method clustered based on super-pixel | |
| CN110070074A (en) | A method of building pedestrian detection model | |
| CN103226835B (en) | Based on method for tracking target and the system of online initialization gradient enhancement regression tree | |
| CN105536205A (en) | Upper limb training system based on monocular video human body action sensing | |
| CN110232308A (en) | Robot gesture track recognizing method is followed based on what hand speed and track were distributed | |
| CN103426025B (en) | A kind of contactless push-up method of counting based on intelligent mobile phone platform | |
| CN104899575A (en) | Human body assembly dividing method based on face detection and key point positioning | |
| CN105893936A (en) | Behavior identification method base on fusion of HOIRM and local feature | |
| CN108764054A (en) | The method that machine person writing's calligraphy of network is fought based on production | |
| CN107516127A (en) | Service robot independently obtains people and wears the method and system for taking article ownership semanteme | |
| Zhang et al. | A kinect based golf swing score and grade system using gmm and svm | |
| CN109508740A (en) | Object hardness identification method based on Gaussian mixed noise production confrontation network | |
| CN106778639A (en) | A kind of exercise data search method based on the description of attitude relative space-time characteristic statisticses | |
| CN103440471A (en) | Human body action identifying method based on lower-rank representation | |
| Li et al. | [Retracted] Deep Learning Algorithm‐Based Target Detection and Fine Localization of Technical Features in Basketball | |
| CN105678348B (en) | A kind of handwritten Chinese character normalization evaluation method and system | |
| CN107644203A (en) | A kind of feature point detecting method of form adaptive classification | |
| CN105844204A (en) | Method and device for recognizing behavior of human body | |
| CN108021883A (en) | The method, apparatus and storage medium of sphere recognizing model of movement | |
| CN103020631B (en) | Human movement identification method based on star model |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210712 Address after: 518000 whole building of Jindi tennis garden sports club at the intersection of Xiangmihu road and Qiaoxiang Road, Futian District, Shenzhen City, Guangdong Province Patentee after: Shenzhen hongjindi sports intelligence Co.,Ltd. Address before: 518000 Jindi Tennis Center, intersection of Xiangmihu road and Qiaoxiang Road, Xiangmihu street, Futian District, Shenzhen City, Guangdong Province Patentee before: SHENZHEN YINGSHI SPORTS TECHNOLOGY Co.,Ltd. |
|
| TR01 | Transfer of patent right |