CN110502564A - Motion characteristic data library generating method, search method and terminal based on posture base - Google Patents
Motion characteristic data library generating method, search method and terminal based on posture base Download PDFInfo
- Publication number
- CN110502564A CN110502564A CN201910799488.1A CN201910799488A CN110502564A CN 110502564 A CN110502564 A CN 110502564A CN 201910799488 A CN201910799488 A CN 201910799488A CN 110502564 A CN110502564 A CN 110502564A
- Authority
- CN
- China
- Prior art keywords
- posture
- motion characteristic
- artis
- base
- characteristic data
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
- G06F16/2228—Indexing structures
- G06F16/2246—Trees, e.g. B+trees
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/25—Integrating or interfacing systems involving database management systems
- G06F16/258—Data format conversion from or to a database
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Processing Or Creating Images (AREA)
Abstract
The present invention relates to a kind of motion characteristic data library generating method, search method and terminals based on posture base, data library generating method is the following steps are included: obtain multiple training data sets of the user under different motion state, wherein, each training data set includes multiple trained acquisition times and training data corresponding with each trained acquisition time;Corresponding posture of each trained acquisition time is obtained based on the training data set, skeleton key artis or crucial limb segment is chosen, is encoded according to the subspace position in his father's artis local coordinate system, be converted into corresponding posture base;Data set and its corresponding digital coding are stored, the motion characteristic data library based on posture base is formed.Human body attitude geometrical characteristic is converted to the digital coding of skeleton key artis by the present invention, while realizing the accurate exercise data retrieval based on content, improves effectiveness of retrieval.
Description
Technical field
It is the present invention relates to a kind of data library generating method, search method and terminal, in particular to a kind of based on posture base
Motion characteristic data library generating method, search method and terminal.
Background technique
As the rise of capturing movement technology and all kinds of optics, work capture the progress of equipment, present people
Have been able to a large amount of human action three-dimensional data files of quick obtaining.Since human action three-dimensional data file can more precisely
The total movement track recorded in experimenter's each period, human body can be obtained by capturing resulting data by analysis movement
The elaboration of movement greatly improves the convenience of action data related work acquisition and the reliability of data.Move number
According to multiplexing and the movement capturing technology that is established as of extensive motion database provide a kind of more time saving and economic scheme,
Meanwhile also to the tissue of motion database and search technique, more stringent requirements are proposed.
The tissue and search technique of motion database are a key technologies for realizing movement capturing data multiplexing.Human body is dynamic
It is a kind of typical higher-dimension time series as sequence, the processing for high dimensional information will if carrying out retrieval using conventional method
A large amount of runing time and memory headroom are expended, and is difficult accurately to retrieve required data based on retrieval content.Therefore,
Choosing suitable character representation method makes retrieval rate and retrieval quality that can receive to seem extremely important.It is existing to movement
The retrieval of database mainly realized by extracting geometrical characteristic and calculating Euclidean distance, and its purpose is to realize pair
The retrieval of motion database content, but calculating to complex geometry feature and Euclidean distance so that when database retrieval need to consume
Take a large amount of runing time, is unable to satisfy real-time demand.
Summary of the invention
In view of the above-mentioned deficiencies in the prior art, it is an object of the present invention to provide a kind of motion characteristic data library based on posture base
Human body attitude geometrical characteristic is converted to digital coding by generation method, search method and terminal, is being realized based on the accurate of content
While exercise data is retrieved, effectiveness of retrieval can be improved.
In order to achieve the above objectives, the present invention provides a kind of motion characteristic data library generating method based on posture base, packets
Include following steps:
Obtain multiple training data sets of the user under different motion state, wherein each training number
It include multiple trained acquisition times and training data corresponding with each trained acquisition time according to collection;
Corresponding posture of each trained acquisition time is obtained based on the training data set, skeleton is chosen and closes
Key artis or crucial limb segment, encode according to the subspace position in his father's artis local coordinate system, are converted into
Corresponding posture base;
Data set and its corresponding digital coding are stored, the motion characteristic data library based on posture base is formed.
Preferably, the method that subspace position of the basis in his father's artis local coordinate system is encoded are as follows:
Father's artis local coordinate system is divided into multiple subspaces according to the size around each reference axis rotation angle, is
Every sub-spaces distribute a digital coding.
Preferably, when dividing subspace in father's artis local coordinate system, only consider the space model for having freedom of movement
It encloses.
Preferably, when dividing subspace in father's artis local coordinate system, so that the boundary of each adjacent subspace has
There is overlapping region.
Preferably, index tree is established according to posture base coding, the corresponding artis of the n omicronn-leaf child node of index tree is at it
Subspace in father's artis local coordinate system is position encoded, and leaf node contains an index structure inode, records as follows
Index information: { posture base coding, initial frame number, place filename, the pointer of previous inode, the finger of next inode
Needle }.
The present invention also provides a kind of action retrieval methods based on posture base, comprising the following steps:
Obtain the exercise data collection of user, wherein the exercise data collection includes multiple acquisition times and each described
Acquisition time corresponding exercise data;
Corresponding posture of each acquisition time is obtained based on the exercise data collection, and posture is converted into corresponding to
Posture base coding;
Posture base is retrieved in motion characteristic data library encodes identical data.
The present invention also provides a kind of terminal, including processor, input equipment, output equipment and memory, the storages
Device is configured for operation said program code, executes the motion feature number for storing program code, the processor
According to the generation method in library.
The present invention also provides a kind of terminal, including processor, input equipment, output equipment and memory, the storages
Device is configured for operation said program code, executes the action retrieval side for storing program code, the processor
Method.
The present invention also provides a kind of computer readable storage medium, the computer storage medium is stored with program and refers to
It enables, the generation side in the motion characteristic data library that described program instruction makes the processor execution described when being executed by a processor
Method.
The present invention also provides a kind of computer readable storage medium, the computer storage medium is stored with program and refers to
It enables, described program instruction makes the processor execute the action retrieval method when being executed by a processor.
Beneficial effect
Motion characteristic data library generating method, search method and terminal proposed by the present invention based on posture base, by human body
Attitude geometry Feature Conversion is the digital coding of skeleton key artis, is realizing the accurate exercise data inspection based on content
While rope, effectiveness of retrieval is improved.
Detailed description of the invention
Fig. 1 is the flow chart of motion characteristic data library generating method;
Fig. 2 is using the joint Hips as the skeleton hierarchical model of root node.
Fig. 3 is the crucial artis and limb segment schematic diagram that embodiment 1 is chosen.
Fig. 4 is the substantially planar and basic axis schematic diagram of human motion.
Fig. 5 is artis Subspace partition schematic diagram.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to embodiments, to the present invention
It is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not used to
Limit the present invention.
Traditional Motion Capture equipment acquires human body movement data, these data generally include multiple time points, often
The corresponding frame of a time point, each frame recording is the exercise data of each artis of human body, such as is displaced, rotary angle information.
There are father and son's node relationships, such as Fig. 2 to illustrate the skeleton hierarchical model using Hips as root node between artis.Pass through
Position of each artis of each frame under world coordinate system can be calculated in the data of set membership and each frame between node
It sets namely the corresponding human body attitude of available each frame.
The tissue and search technique of motion database are a key technologies for realizing movement capturing data multiplexing.Existing skill
It is usually all by extracting geometrical characteristic and calculating Euclidean distance to realize the retrieval to motion database content in art
Come what is realized.Every time when retrieval, the Euclidean distance of each frame in the movement retrieved needed for calculating and database is required, so that inspection
Rope is inefficient.
Idea of the invention is that a kind of motion characteristic data library generating method and action retrieval method are provided, by human body attitude
Geometrical characteristic is converted to digital coding, and while realizing the accurate exercise data retrieval based on content, retrieval can be improved
Efficiency.
Embodiment 1 provides a kind of motion characteristic data library generating method based on posture base, comprising the following steps:
S1: multiple training data sets of the user under different motion state are obtained, wherein each training
Data set includes multiple trained acquisition times and training data corresponding with each trained acquisition time;
S2: corresponding posture of each trained acquisition time is obtained based on the training data set, chooses human body bone
Bone key artis or crucial limb segment, encode according to the subspace position in his father's artis local coordinate system, turn
Change corresponding posture base into;
S3: data set and its corresponding digital coding are stored, and form the motion characteristic data library based on posture base.
In step S2, human body attitude is obtained from each frame, and human body attitude is converted into posture base digital coding.Implement
Example 1 has chosen 12 artis in skeleton framework model according to the movement characteristic of human body attitude and constitutes 9 limbs altogether
Section, as shown in Figure 3.Solid stain in figure is the 12 crucial artis chosen, 9 be represented by dashed line between 12 artis
A limb segment is the key that the expression human body attitude limb segment that embodiment 1 is chosen, respectively left and right large arm, forearm, thigh, shank
And trunk.There is set memberships between node.
In embodiment 1, using each skeleton key artis as origin, independent local coordinate system is established for them.Limb
Body section is moved around artis, and largest motion range may be constructed one using the artis as the ball in the center of circle in theory
Body.The sphere space is divided into multiple subspaces, the one digital coded representation of every sub-spaces then records the sub- artis
Or subspace locating for limb segment, so that it may indicate position and the posture of the sub- artis or limb segment.All key artis
Coded combination, so that it may indicate the posture of human body, this coded combination is known as posture base by us.
In human body movement data processing, usually the substantially planar of human body can be defined as horizontal plane, frontal plane, sagittal
Three, face plane, as shown in figure 4, wherein horizontal plane is the section that crosscutting body is parallel to the ground under upright state, by body point
For upper and lower two parts;Frontal plane is that human body is divided into forward and backward two parts using body bilateral diameter as longitudal section made by tangent line;Arrow
Shape face is that human body is divided into left and right two parts using body anteroposterior diameter as longitudal section made by tangent line;The basic axis of human body is defined
For frontal axis, vertical axis, sagittal axis, wherein frontal axis is X-axis, and perpendicular to sagittal plane, direction is left and right direction, be frontal plane with
The intersection of horizontal plane;Vertical axis is Y-axis, and perpendicular to horizontal plane, it is the intersection of frontal plane and sagittal plane that direction, which is upper and lower direction,;
Sagittal axis is Z axis, and perpendicular to frontal plane, it is the intersection of sagittal plane and horizontal plane that direction, which is front and back direction,.
Attached drawing 5 is the schematic diagram of 1 Subspace partition of embodiment.Embodiment 1 presses the space of father's artis local coordinate system
It is divided according to the angle rotated around X-axis, around Y-axis, about the z axis, as shown in Figure 5.
In actual motion, what the freedom degree of limb segment movement was often limited by, therefore its largest motion range is simultaneously
It is not a sphere.Such as the freedom degree that shank has around knee joint and only one is rotated around X-axis, maximum rotating range are
180°.Embodiment 1 only considers the sky that its practical motion range is constituted for each limb segment when carrying out posture base coding
Between.
It, can be by the movement model around the angle that X rotates according to limb segment if the freedom degree that limb segment is only rotated around X-axis
It encloses and is divided into 3 sub-spaces, indicated respectively with number 0,1,2, corresponding angular range is 0-60 degree, 60-120 degree, 120-180
Degree, as shown in the side view a of Fig. 5.If limb segment is 0-180 degree around the rotary freedom of X-axis and Z axis, according to limbs
The motion range can be divided into 9 sub-spaces by the angle that section is rotated around X angle rotate and about the z axis, use digital 00 respectively,
01, it 02,10,11,12,20,21,22 indicates, the corresponding angle rotated around X-axis of the first position digital coding, second-order digit coding
The corresponding angle rotated about the z axis, as shown in Fig. 5 top view b.
As shown in figure 5, in order to improve recall ratio, embodiment 1 is when dividing adjacent subspace, so that each adjacent subspace
Boundary have size be δ overlapping region, that is, increase the angular range of every sub-spaces.Due to the human body bone in overlay region
There are two different codings for bone artis tool, this can enable same frame data by multiple posture base coded representations, therefore centainly encode
The movement can be retrieved in range, recall ratio can be made to get a promotion in action retrieval.
Geometrical characteristic, which is converted to digital coding, can greatly improve the efficiency of database retrieval.In order to which restriction will be searched for
In in effective data area, embodiment 1 is encoded based on posture base, for motion characteristic Database index tree.Establish rope
Following two step can be performed in the process for drawing tree: (1) carrying out posture base coding to each frame data;(2) identical movement piece will be encoded
Section is inserted into corresponding leaf node in index tree.The corresponding artis of the n omicronn-leaf child node of index tree is in his father's artis part
Subspace in coordinate system is position encoded, and leaf node contains an index structure inode, records following index information: { appearance
State base coding, initial frame number, place filename, the pointer of previous inode, the pointer of next inode }.
According to the index tree constructing method of embodiment 1, each leaf node is connected to the identical action movie of a coding
The set of section, all these set constitute entire motion database.Exercise data stream passes through code tree encoding filter, is divided
At the motion segments of suitable length.It is this to be promoted the comparison of movement to segment one by frame level-one based on the segmented mode of feature
Grade, greatly accelerates the execution efficiency of searching algorithm, while ensure that the time continuity of movement.The foundation of index increases pre-
Calculation amount, but for same database, index only needs to establish once, when database size expands, it can be achieved that the line of index
Property extension.
Embodiment 2 realizes a kind of action retrieval method based on posture base, comprising the following steps:
Obtain the exercise data collection of user, wherein the exercise data collection includes multiple acquisition times and each described
Acquisition time corresponding exercise data;
Corresponding posture of each acquisition time is obtained based on the exercise data collection, and posture is converted into corresponding to
Posture base coding;
Posture base is retrieved in the motion characteristic data library established according to the method described above encodes identical data.
The present invention also provides a kind of terminal, including processor, input equipment, output equipment and memory, the storages
Device is configured for operation said program code, executes the motion feature number for storing program code, the processor
According to the generation method in library.
The present invention also provides a kind of terminal, including processor, input equipment, output equipment and memory, the storages
Device is configured for operation said program code, executes the action retrieval side for storing program code, the processor
Method.
The present invention also provides a kind of computer readable storage medium, the computer storage medium is stored with program and refers to
It enables, the generation side in the motion characteristic data library that described program instruction makes the processor execution described when being executed by a processor
Method.
The present invention also provides a kind of computer readable storage medium, the computer storage medium is stored with program and refers to
It enables, described program instruction makes the processor execute the action retrieval method when being executed by a processor.
Although the embodiments of the invention are described in conjunction with the attached drawings, but those skilled in the art can not depart from this hair
Various modifications and variations are made in the case where bright spirit and scope, such modifications and variations each fall within the claims in the present invention
Within limited range.
Claims (10)
1. a kind of motion characteristic data library generating method based on posture base, which comprises the following steps:
Obtain multiple training data sets of the user under different motion state, wherein each training data set
Including multiple trained acquisition times and training data corresponding with each trained acquisition time;
Corresponding posture of each trained acquisition time is obtained based on the training data set, skeleton key is chosen and closes
Node or limb segment encode according to the subspace position in his father's artis local coordinate system, are converted into corresponding appearance
State base;
Data set and its corresponding digital coding are stored, the motion characteristic data library based on posture base is formed.
2. based on the motion characteristic data library generating method described in claim 1 based on posture base, which is characterized in that described
The method encoded according to the subspace position in his father's artis local coordinate system are as follows:
The subspace of father's artis local coordinate system is divided according to the rotation angle around each reference axis, is every sub-spaces
Distribute a digital coding.
3. based on the motion characteristic data library generating method as claimed in claim 2 based on posture base, which is characterized in that closed in father
When dividing subspace in node local coordinate system, the spatial dimension for having freedom of movement is only considered.
4. based on the motion characteristic data library generating method as claimed in claim 2 based on posture base, which is characterized in that closed in father
When dividing subspace in node local coordinate system, so that the boundary of each adjacent subspace has overlapping region.
5. based on the motion characteristic data library generating method as claimed in claim 2 based on posture base, which is characterized in that according to appearance
State base coding establishes index tree, and the corresponding artis of the n omicronn-leaf child node of index tree is in his father's artis local coordinate system
Subspace is position encoded, and leaf node contains an index structure inode, records following index information: posture base coding,
Initial frame number, place filename, the pointer of previous inode, the pointer of next inode }.
6. a kind of action retrieval method based on posture base, which comprises the following steps:
Obtain the exercise data collection of user, wherein the exercise data collection includes multiple acquisition times and each acquisition
Time point corresponding exercise data;
Corresponding posture of each acquisition time is obtained based on the exercise data collection, and posture is converted into corresponding appearance
State base coding;
Posture base is retrieved in the motion characteristic data library generated according to any one of claim 1-5 method encodes identical number
According to.
7. a kind of terminal, including processor, input equipment, output equipment and memory, which is characterized in that the memory is used for
Program code is stored, the processor is configured for operation said program code, executes as described in claim any one of 1-5
Motion characteristic data library generation method.
8. a kind of terminal, including processor, input equipment, output equipment and memory, which is characterized in that the memory is used for
Program code is stored, the processor is configured for operation said program code, executes movement inspection as claimed in claim 6
Suo Fangfa.
9. a kind of computer readable storage medium, which is characterized in that the computer storage medium is stored with program instruction, described
Program instruction makes the processor execute motion characteristic data as described in any one in claim 1-5 when being executed by a processor
The generation method in library.
10. a kind of computer readable storage medium, which is characterized in that the computer storage medium is stored with program instruction, institute
Stating program instruction when being executed by a processor makes the processor execute action retrieval method as claimed in claim 6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910799488.1A CN110502564A (en) | 2019-08-28 | 2019-08-28 | Motion characteristic data library generating method, search method and terminal based on posture base |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910799488.1A CN110502564A (en) | 2019-08-28 | 2019-08-28 | Motion characteristic data library generating method, search method and terminal based on posture base |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110502564A true CN110502564A (en) | 2019-11-26 |
Family
ID=68589978
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910799488.1A Pending CN110502564A (en) | 2019-08-28 | 2019-08-28 | Motion characteristic data library generating method, search method and terminal based on posture base |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110502564A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111754619A (en) * | 2020-06-29 | 2020-10-09 | 武汉市东旅科技有限公司 | Bone space data acquisition method, acquisition device, electronic device and storage medium |
CN113987285A (en) * | 2021-12-27 | 2022-01-28 | 北京理工大学 | Hidden state-based motion characteristic database generation method and search method |
CN116189309A (en) * | 2022-11-16 | 2023-05-30 | 北京理工大学 | Method for processing human motion data and electronic equipment |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103886588A (en) * | 2014-02-26 | 2014-06-25 | 浙江大学 | Feature extraction method of three-dimensional human body posture projection |
CN105389539A (en) * | 2015-10-15 | 2016-03-09 | 电子科技大学 | Three-dimensional gesture estimation method and three-dimensional gesture estimation system based on depth data |
CN109949368A (en) * | 2019-03-14 | 2019-06-28 | 郑州大学 | A kind of human body three-dimensional Attitude estimation method based on image retrieval |
-
2019
- 2019-08-28 CN CN201910799488.1A patent/CN110502564A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103886588A (en) * | 2014-02-26 | 2014-06-25 | 浙江大学 | Feature extraction method of three-dimensional human body posture projection |
CN105389539A (en) * | 2015-10-15 | 2016-03-09 | 电子科技大学 | Three-dimensional gesture estimation method and three-dimensional gesture estimation system based on depth data |
CN109949368A (en) * | 2019-03-14 | 2019-06-28 | 郑州大学 | A kind of human body three-dimensional Attitude estimation method based on image retrieval |
Non-Patent Citations (2)
Title |
---|
刘云根等: "基于人体姿势编码的运动数据检索", 《计算机辅助设计与图形学学报》 * |
黄天羽等: "一种基于几何特征编码的运动搜索", 《***仿真学报》 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111754619A (en) * | 2020-06-29 | 2020-10-09 | 武汉市东旅科技有限公司 | Bone space data acquisition method, acquisition device, electronic device and storage medium |
CN113987285A (en) * | 2021-12-27 | 2022-01-28 | 北京理工大学 | Hidden state-based motion characteristic database generation method and search method |
CN116189309A (en) * | 2022-11-16 | 2023-05-30 | 北京理工大学 | Method for processing human motion data and electronic equipment |
CN116189310A (en) * | 2022-11-16 | 2023-05-30 | 北京理工大学 | Method for providing human motion data set and electronic equipment |
CN116469159A (en) * | 2022-11-16 | 2023-07-21 | 北京理工大学 | Method for acquiring human motion data and electronic equipment |
CN116469159B (en) * | 2022-11-16 | 2023-11-14 | 北京理工大学 | Method for acquiring human motion data and electronic equipment |
CN116189310B (en) * | 2022-11-16 | 2024-01-23 | 北京理工大学 | Method for providing human motion data set and electronic equipment |
CN116189309B (en) * | 2022-11-16 | 2024-01-30 | 北京理工大学 | Method for processing human motion data and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Punnakkal et al. | BABEL: Bodies, action and behavior with english labels | |
CN110502564A (en) | Motion characteristic data library generating method, search method and terminal based on posture base | |
CN110222580A (en) | A kind of manpower 3 d pose estimation method and device based on three-dimensional point cloud | |
CN110188700B (en) | Human body three-dimensional joint point prediction method based on grouping regression model | |
CN105574510A (en) | Gait identification method and device | |
CN102508867A (en) | Human-motion diagram searching method | |
CN113987285B (en) | Hidden state-based motion characteristic database generation method and search method | |
CN104268285A (en) | Motion capture database searching method based on dynamic bending algorithm | |
Zhu et al. | Mocanet: Motion retargeting in-the-wild via canonicalization networks | |
CN110516114A (en) | Motion database automatic marking method and terminal based on posture base | |
CN101241601A (en) | Graphic processing joint center parameter estimation method | |
CN110287912A (en) | Method, apparatus and medium are determined based on the target object affective state of deep learning | |
CN110516112A (en) | A kind of human action search method and equipment based on hierarchical model | |
CN102761759A (en) | Motion coding image and generation module, image processing module and motion display module | |
CN1567383A (en) | Method for three-dimensional animation production using literal script | |
Hu et al. | Learning Explicit Contact for Implicit Reconstruction of Hand-Held Objects from Monocular Images | |
Li et al. | An attention-seq2seq model based on CRNN encoding for automatic Labanotation generation from motion capture data | |
Pan et al. | Combining pose and trajectory for skeleton based action recognition using two-stream RNN | |
Ye et al. | Semantics2Hands: Transferring Hand Motion Semantics between Avatars | |
Chen et al. | Adaptive reconstruction of human motion on wireless body sensor networks | |
Wang et al. | Design of calligraphy aesthetic evaluation model based on deep learning and writing action | |
Tang et al. | PAMS-DP: Building a Unified Open PAMS Human Movement Data Platform | |
Ugolotti et al. | Differential evolution based human body pose estimation from point clouds | |
Zan et al. | A framework for human motion segmentation based on multiple information of motion data | |
Zhou et al. | Motion balance ability detection based on video analysis in virtual reality environment |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20191126 |
|
WD01 | Invention patent application deemed withdrawn after publication |