CN101154289A - Method for tracing three-dimensional human body movement based on multi-camera - Google Patents
Method for tracing three-dimensional human body movement based on multi-camera Download PDFInfo
- Publication number
- CN101154289A CN101154289A CNA2007100442191A CN200710044219A CN101154289A CN 101154289 A CN101154289 A CN 101154289A CN A2007100442191 A CNA2007100442191 A CN A2007100442191A CN 200710044219 A CN200710044219 A CN 200710044219A CN 101154289 A CN101154289 A CN 101154289A
- Authority
- CN
- China
- Prior art keywords
- skeleton
- human
- frame
- tracks
- bone
- 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
Images
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Image Analysis (AREA)
Abstract
A three-dimensional human body motion trace method based on a multi-camera in the computer vision technical field has the following steps: firstly, tissue data points on the human body surface are picked up and outmost points are obtained through processing the tissue data; secondly, a standard three-dimensional skeleton model which is used for dynamic posture adjustment of the model during the trace process is established before skeleton pick-up and human body motion trace, thereby realizing fitting of the model and an original three-dimensional image; thirdly, a skeleton observation method is established and the skeleton model posture is adjusted according to the status that whether the number of the tissue points in each section of bone sleeve is maximal; fourthly, an implementation approach of optimal matching skeleton is established and the state that whether the current skeleton pose has met the trace requirements or not is judged through changing the space positions of root nodes and the angle of rotation of each boot section winding around the father node. The invention can quickly get the human body three-dimensional skeleton based on the existing human body tissue data, thereby realizing the human body motion trace.
Description
Technical field
The present invention relates to the method in a kind of telecommunication technology field, specifically is a kind of method of following the tracks of based on the 3 d human motion of many orders camera.
Background technology
It is the focus and the difficult point of current computer vision area research that 3 d human motion is followed the tracks of.Extracting the human skeleton model in the middle of the video data is an effectively monitoring and tracking.The human skeleton main application fields that obtains from the three-dimensional voxel data has: supervisory system, virtual reality, sophisticated user interface, intelligent environment, amusement, motion analysis, medical science, education etc.The three-dimensional voxel English word is voxel, it is the abbreviation of two words of Volume Element, Voxel is called " voxel ", it just is equivalent to " pixel " in the two dimensional image--" pixel ", it is the basic blockage in the three dimensions, X, Y and three information such as coordinate of Z of comprising each point are obtained three-dimensional human body voxel data by many orders camera and are carried out that attitude is estimated and the human motion tracking is a kind of very novel method.Yet the voxel data that 3 D human body has only been arranged can't be enough to solve many practical problemss.The voxel data that obtains by many orders camera is directly rebuild and to its tracking, because the difference of camera quantity causes the result uneven, data volume is very big on the whole, and arithmetic speed is slow and be difficult for carrying out data in real time transmission and Web publishing, does not reach the effect of following the tracks of and monitoring.Existing two-dimension human body motion tracking is followed the tracks of in human motion, also has 3 d human motion to follow the tracks of.The 3 d human motion tracking is more complex with respect to two dimension.It is very general to analyze human motion by voxel data, thereby but obtaining accurately human skeleton by voxel data follows the tracks of but to human motion that few people relate to.This work meaning is more special, both the tracking results that obtains can be used for supervisory system, can be used as the identification that intermediate result is carried out human posture and action again, does further research.Generally speaking, the accurate tracking of human motion is significant, is the difficult problem of whole computer vision field.
Find by prior art documents, Caillette, F. the paper " Real-time markerlesshuman body tracking using colored voxels and 3D blobs " that waits the people in " Third IEEE andACM International Symposium on Mixed and Augmented Reality " (the 3rd international symposium of IEEE and ACM), to deliver about mixing and expansion reality in 2004, come human body is followed the tracks of with the voxel information that has color in (with colored voxel and the unmarked human body of three-dimensional module real-time follow-up) paper, this method needs tracked target and scene to have reasonable color contrast poor, and this has just limited its application in the middle of engineering greatly.
Summary of the invention
The objective of the invention is to overcome deficiency of the prior art, a kind of method of following the tracks of based on the 3 d human motion of many orders camera is provided, make it have good stability, rapidity is good, the accuracy height, and output file is little, is easy to the characteristics of preserving and transmitting.The present invention is by the tracking to human motion, judge and predict attitude, action of human body etc. exactly, thereby can allow computing machine analyze voluntarily and judge, in the middle of a supervisory system, this information can be sent to monitor terminal in real time, thereby corresponding measure is also in time taked in unusual circumstance in real time.
The present invention is achieved by the following technical solutions, may further comprise the steps:
The first, handle voxel data at the number of voxels strong point of extracting human body surface, obtains outermost point;
The second, extracting the three-dimensional human skeleton model that skeleton makes up a standard before human motion is followed the tracks of, be used in the middle of the process of following the tracks of, it being carried out dynamic stance adjustment, to realize the match of it and original three-dimensional image;
The 3rd, make up the human skeleton abstracting method, whether be the maximum skeleton pattern attitude of adjusting according to the tissue points number in every section bone sleeve in the three-dimensional human skeleton model;
The 4th, the implementation method of structure Optimum Matching skeleton judges around the anglec of rotation of its father node whether current human skeleton attitude has reached tracer request by locus and every section bone of changing root node in the three-dimensional human skeleton model.
The number of voxels strong point of described extraction human body surface, voxel data is handled, obtained outermost point, be meant: adopt 16 video cameras to obtain after the 3 D human body voxel data, in order to reduce data, to promote processing speed, the outmost voxel of human body is sought out.The three-dimensional voxel data have been arranged, all voxels have all just been shown with computing machine can obtain 3-D view.In order to obtain the outermost point of 3-D view, the present invention has designed 3 * 3 * 3 image template.Wherein 3 * 3 * 3 be meant that the front, back, left, right, up, down six direction of each voxel (comprising itself) in three dimensions forms 27 points that amount to of cube template.Be different from two dimensional image, this template is 26 connected domains.For any one voxel, up, down, left, right, before and after and its Euclidean distance are less than or equal to
Point all to can be regarded as be this neighborhood of a point point, amount to 26.Judging in the middle of deterministic process whether this neighborhood of a point point has powerful connections a little, then is the outermost layer point if having.
The three-dimensional human skeleton model of a standard of described structure is meant: carrying out before skeleton extract follows the tracks of human motion the three-dimensional human skeleton model of a standard of design.In the middle of this skeleton pattern, the definition plurality of nodes defines a root node, controls the space rotation and the spatial translation of whole skeleton.Father node of each node definition, every section skeleton is all done rotation and translation around its father node in local coordinate system, can draw of the change of each node at last with respect to the locus of root node, thereby can determine each node at three-dimensional coordinate, and then determine the skeleton posture of every frame.For every section bone, it has several parameters to determine, for example ID number, father node, length etc.Every bone defines unique ID number, present node and father node can be coupled together by ID number; The length of every section bone is all inequality, can obtain by setting or machine learning.
Described structure human skeleton abstracting method is meant: judge the effect that human motion is followed the tracks of by the matching degree of manikin and skeleton observational characteristic.Tracking results by preceding frame is also utilized the locus of this frame root node and each section bone is followed the tracks of down frame around the anglec of rotation parameter of its father node human motion.The parameter of first frame is obtained by man-machine interaction or machine learning.Because the gradually changeable of human motion, front and back two frame parameters change usually within the specific limits, and this scope can artificially be specified according to general knowledge.But taked a simple efficient ways: give every section bone cover a cylinder, the number of the tissue points of calculating in sleeve, as long as it reach maximum and the skeleton that obtains in body surface's point, then stop to calculate, think that this skeleton posture is optimum skeleton posture, has reached the purpose of following the tracks of human motion by the skeleton posture.
The implementation method of described structure Optimum Matching skeleton, be meant: for the three-dimensional voxel that makes human skeleton and every frame reaches optimum matching, determine the definite locus of each node of every frame skeleton, thus the three-dimensional coordinate of the root node of necessary controlled whole skeleton translation and rotation.After obtaining the first frame three-dimensional human skeleton, next frame skeleton attitude is adjusted as benchmark with the former frame skeleton.The parameter that changes every frame skeleton posture mainly is that every bone is done rotational transform around X, Y, Z coordinate axis respectively around its father node place local coordinate system, in the middle of the process of this conversion, adopt the probability genetic algorithm to come every bone is carried out iteration around the angle of its father node, up to the human skeleton that obtains human body with interior and can represent human body attitude till.Wherein every section bone has individual local optimum, and whole skeleton posture reached optimum when all bone all reached local optimum, and to the tracking end of this frame, the tracking that enters next frame is to the last till the frame end.
Compared with prior art, the present invention is simply effective, and its key is to obtain rapidly the three-dimensional framework of human body on the basis of existing human body voxel data, thereby the motion posture that can very clearly judge human body is to follow the tracks of human motion.Utilize the present invention to carry out human motion and follow the tracks of, both can be used as the result and utilized, can be used as the intermediate result of next step pattern-recognition again.Because the data volume little (less than 1K) of every frame human skeleton file, and the size of every frame 3 D image file very big (more than the 5M), so the human skeleton file that obtains can be saved storage space greatly, save network central transmission time and transmission cost, be specially adapted in the middle of the supervisory system, simultaneously also can be applied to computer animation, recreation, fields such as virtual reality.
Description of drawings
Fig. 1 is the locus synoptic diagram of embodiment of the invention camera and object
Fig. 2 is the embodiment of the invention 3 * 3 * 3 image template synoptic diagram
Fig. 3 is an embodiment of the invention 3 D human body voxel image picture group, and wherein: from left to right every image name is respectively the 15th frame, the 45th frame, the 75th frame, the 105th frame, the 135th frame, the 165th frame.
Fig. 4 is embodiment of the invention standard skeleton pattern figure
Fig. 5 is an embodiment of the invention skeleton sleeve synoptic diagram
Fig. 6 is an embodiment of the invention picture group (only skeleton) as a result
Wherein: from left to right every image name is respectively the 15th frame, the 45th frame, the 75th frame, the 105th frame, the 135th frame, the 165th frame.
Fig. 7 embodiment is picture group (skeleton and voxel) as a result
Wherein: from left to right every image name is respectively the 15th frame, the 45th frame, the 75th frame, the 105th frame, the 135th frame, the 165th frame.
Embodiment
Below in conjunction with accompanying drawing embodiments of the invention are elaborated: present embodiment has provided detailed embodiment and process being to implement under the prerequisite with the technical solution of the present invention, but protection scope of the present invention is not limited to following embodiment.
1. extract the human body surface voxel.As shown in Figure 1, arrange 16 video cameras in the middle of a scene, camera is two-layer up and down among the figure, obtains the two dimensional image of human body and makes up 3-D view by these images by each camera, and each primitive of forming 3-D view is exactly a voxel.In order to reduce the computational expense of computing machine, improve the entire method processing speed, reduce the processing time, adopt 3 * 3 * 3 templates as shown in Figure 2 that the data point of every frame is handled, the current voxel that will detect of solid circles representative among the figure, empty circles is represented 26 adjacent voxels altogether around it.Whole video amounts to 191 frames, and the image that outermost layer voxel is as shown in Figure 3 formed is for getting the image of a frame every 30 frames since the 15th frame.In order to show that more clearly result of implementation, present embodiment have adopted the mode of getting a point every ten points to show raw data.Concrete steps are as follows:
(1) earlier every frame three-dimensional voxel data is carried out complete scan one time, find out the maximal value and the minimum value of coordinate of the point of X, Y, Z direction respectively.
(2) according to maximal value and the minimum value obtained, the maximal value of each direction is subtracted each other and is obtained M, N, P respectively, sets up the 3D grid space of a M * N * P dimension, has the place of voxel to put 1, otherwise puts 0.
(3) utilize 3 * 3 * 3 templates that every frame element is handled, the value of non-outermost point is put 0 by 1, thereby obtain impact point.
2. make up standard human body skeleton pattern
(1) determines the human skeleton essential information.Be illustrated in figure 4 as the standard skeleton of the embodiment of the invention, it is made up of some bones, and the length of every bone is determined different length respectively according to different people, marks ID number of good every section skeleton, has determined the mutual relationship between the node.
(2) in the middle of this standard skeleton pattern, the definition plurality of nodes.Wherein root node is being controlled whole skeleton in rotation of the space of world coordinate system and spatial translation.Every section skeleton is all done rotation and translation around its father node in local coordinate system, can obtain the change of each node with respect to the locus of root node at last.For around the transformation matrix that is rotated of axle that does not overlap with coordinate axis, utilize the rotation of translation and coordinate axis compound and obtain.At general three-dimensional rotation situation: the translation skeleton makes it overlap with a coordinate axis that is parallel to this skeleton; Finish the rotation of appointment for this; Translation of object is moved back into original position with skeleton.Every section bone can be finished the adjustment of attitude around the translation and the rotation of father node, thereby reaches the purpose of tracking.
3. make up the human skeleton abstracting method
In the process of the matching degree of manikin and skeleton observational characteristic, the tracking results by preceding frame is also utilized the locus of this frame root node and each section bone is followed the tracks of down frame around these parameters of the anglec of rotation of its father node human motion.The parameter of first frame is obtained by man-machine interaction or machine learning.By the probability genetic algorithm, set these two parameters of population scale number and iterations after, the skeleton of former frame carries out computing as the reference of back frame prime number certificate, can obtain the skeleton of next frame apace, thereby human motion is followed the tracks of.As shown in Figure 5, give every section bone cover a cylinder, suppose the always total N root skeleton of human skeleton, then will overlap N cylinder to skeleton, and calculate the number of the tissue points in every section skeleton cylinder, the skeleton observation function of definition is as follows:
K represents the bone number that defines, and N represents the quantity of Voxel data mid point,
S=[s
1, s
2..., s
k], s represents whole skeleton, s
iExpression bone,
U (s
i) the interior space of expression skeleton cylinder.
In the process that is optimized, when the number of the three-dimensional point in tube is maximum, just think tracking effect the best this moment.The tracking of t frame is equivalent to and seeks the matrix parameter that makes the skeleton best results in the feasible zone of all matrix parameters, because the higher and common multimodal of the terrain surface skeleton observation function of parameter s dimension and contain a lot of local extremums, the tradition optimized Algorithm is difficult to find global optimum's point, so utilize the probability genetic algorithm that the skeleton observation function is optimized to reach the purpose of tracking.
4. make up the implementation method of Optimum Matching skeleton
Determine the definite locus of each node of every frame skeleton, the 3 d space coordinate of root node can obtain by manual or machine learning.Utilization probability genetic algorithm and in conjunction with the 3 d human motion analysis is carried out compound position coding to the anglec of rotation of three coordinate axis of every section bone father node place coordinate system, and step is as follows:
1) human skeleton is analyzed, determined to carry out the number of the individuality of encoding compound position in conjunction with the limit of people's actual joint motions, can promote processing speed under the situation not influencing as a result like this, save time;
2) to every bone the parameter (angle) that must use carry out compound position coding;
3) determine population scale number and iterations, the population scale number is high more, and iterations is many more, and human skeleton posture and original three-dimensional image are just mated more;
4) produce initial population (each root bone is around the set of the anglec of rotation of corresponding father node) and it is observed, utilize the skeleton observation function to observe the number of the voxel in cylinder;
5) if this quantity increases earlier to tend to be steady again, then change step 6) in the middle of iterative process; Then need to produce new population if increase always, it is rapid to get back to previous step;
6) iteration finishes.
The foregoing description result is as shown in Figure 6 and Figure 7: from left to right every two field picture in from left to right every two field picture corresponding diagram 3 among Fig. 6, be the tracking skeleton that obtains, also from left to right every two field picture in the corresponding diagram 3 of from left to right every two field picture among Fig. 7, but it also shows the tracking skeleton that obtains simultaneously.What take in implementation process is the basic exercise of a people in a three dimensions, comprises normal walking, and stretches out one's hand, and basic form such as turn round.Every two field picture comprises general 45000 three-dimensional point.Result of implementation shows, not only can well obtain framework information under without the situation of voxel colouring information, and has saved the additional computational overhead that increases because using this.Need the general 1 minute time from every frame prime number according to obtaining this frame skeleton data, reach real-time requirement substantially, be better than using other method greatly.Thereby as can be seen the skeleton that obtains of the present invention all at human body with interior and reaction that can the 100% motion posture of human body this moment, can be to next step action recognition, data transmission plays critical effect.
Claims (9)
1. a method of following the tracks of based on the 3 d human motion of many orders camera is characterized in that, may further comprise the steps:
The first, handle voxel data at the number of voxels strong point of extracting human body surface, obtains outermost point;
The second, before being followed the tracks of, human motion makes up the three-dimensional human skeleton model of a standard carrying out skeleton extract, be used in the middle of the process of following the tracks of, it being carried out dynamic stance adjustment, to realize the match of it and original three-dimensional image;
The 3rd, make up the human skeleton abstracting method, whether be the maximum skeleton pattern attitude of adjusting according to the tissue points number in every section bone sleeve in the three-dimensional human skeleton model;
The 4th, the implementation method of structure Optimum Matching skeleton judges around the anglec of rotation of its father node whether current human skeleton attitude has reached tracer request by locus and every section bone of changing root node in the three-dimensional human skeleton model.
2. method of following the tracks of according to claim 1 based on the 3 d human motion of many orders camera, it is characterized in that, the number of voxels strong point of described extraction human body surface, voxel data is handled, obtain outermost point, be meant: obtain the 3 D human body voxel data and obtain 3-D view with the computing machine demonstration by 16 video cameras, adopt 3 * 3 * 3 image template to obtain the outermost point of 3-D view then, wherein 3 * 3 * 3 be meant each voxel in three dimensions, comprise itself before, after, a left side, right, on, following six direction is formed 27 points that amount to of cube template, this template is 26 connected domains, for any one voxel, to last, down, a left side, right, before, back and its Euclidean distance are less than or equal to
Point all to can be regarded as be this neighborhood of a point point, amount to 26, judge in the middle of deterministic process whether this neighborhood of a point point has powerful connections a little, then be the outermost layer point as if having.
3. method of following the tracks of according to claim 2 based on the 3 d human motion of many orders camera, it is characterized in that, the three-dimensional human skeleton model of a standard of described structure, be meant: in the middle of the three-dimensional human skeleton model, definition plurality of nodes and a root node, control the space rotation and the spatial translation of whole skeleton, father node of each node definition, every section skeleton is all done rotation and translation around its father node in local coordinate system, draw of the change of each node at last with respect to the locus of root node, thereby determine each node at three-dimensional coordinate, and then determine the skeleton posture of every frame.
4. method of following the tracks of according to claim 3 based on the 3 d human motion of many orders camera, it is characterized in that, described every section bone, it has several parameters to determine, comprise ID number, father node, length, unique ID number of every bone definition couples together present node and father node by ID number, the length of every section bone is all inequality, obtains by setting or machine learning.
5. method of following the tracks of according to claim 4 based on the 3 d human motion of many orders camera, it is characterized in that, described every section skeleton is all done rotation and translation around its father node in local coordinate system, wherein: for around the transformation matrix that is rotated of axle that does not overlap with coordinate axis, utilize the rotation of translation and coordinate axis compound and obtain; Rotating the peaceful condition of shifting one's love comprises: the translation skeleton makes it overlap with a coordinate axis that is parallel to this skeleton, for this rotation and translation of object of finishing appointment skeleton is moved back into original position.
6. method of following the tracks of according to claim 1 based on the 3 d human motion of many orders camera, it is characterized in that, described structure human skeleton abstracting method, be meant: judge the effect that human motion is followed the tracks of according to the matching degree of manikin and skeleton observational characteristic, tracking results by preceding frame is also utilized the locus of this frame root node and each section bone is followed the tracks of down frame around the anglec of rotation parameter of its father node human motion, the parameter of first frame is obtained by man-machine interaction or machine learning, front and back two frame parameter variation ranges adopt following method to determine: give every section bone cover a cylinder, the number of the tissue points of calculating in sleeve, as long as it reach maximum and the skeleton that obtains in body surface's point, then stop to calculate, think that this skeleton posture is optimum skeleton posture, has reached the purpose of following the tracks of human motion by the skeleton posture.
7. method of following the tracks of according to claim 1 based on the 3 d human motion of many orders camera, it is characterized in that, the implementation method of described structure Optimum Matching skeleton, be meant: after obtaining the first frame three-dimensional human skeleton, next frame skeleton attitude is adjusted as benchmark with the former frame skeleton, the parameter that changes every frame skeleton posture be every bone around its father node place local coordinate system respectively around X, Y, the Z coordinate axis is done rotational transform, in the middle of the process of this conversion, adopt the probability genetic algorithm to come every bone is carried out iteration around the angle of its father node, up to the human skeleton that obtains human body with interior and can represent human body attitude till, wherein every section bone has individual local optimum, whole skeleton posture reached optimum when all bone all reached local optimum, tracking to this frame finishes, enter the tracking of next frame, to the last till the frame end.
8. according to claim 1 or 6 described methods of following the tracks of based on the 3 d human motion of many orders camera, it is characterized in that, the implementation method of described structure Optimum Matching skeleton, be specially: the definite locus of determining each node of every frame skeleton earlier, the 3 d space coordinate of root node obtains by manual or machine learning, uses the probability genetic algorithm that the anglec of rotation of three coordinate axis of every section bone father node place coordinate system is carried out compound position coding then:
1) determines to carry out the number of the individuality of encoding compound position in conjunction with the limit of people's actual joint motions;
2) parameter that will use every bone is that the anglec of rotation of three coordinate axis of every section bone father node place coordinate system is carried out compound position coding;
3) determine population scale number and iterations, the population scale number is high more, and iterations is many more, and human skeleton posture and original three-dimensional image are just mated more;
4) producing initial population is the set of each root bone around the anglec of rotation of corresponding father node, and it is observed, and utilizes the skeleton observation function to observe the number of the voxel in cylinder;
5) if this quantity increases earlier to tend to be steady again, then change step 6) in the middle of iterative process; Then need to produce new population if increase always, it is rapid to get back to previous step;
6) iteration finishes.
9. method of following the tracks of based on the 3 d human motion of many orders camera according to claim 8 is characterized in that, described skeleton observation function is specific as follows:
K represents the bone number that defines, and N represents the quantity of Voxel data mid point,
S=[s
1, s
2..., s
k], s represents whole skeleton, s
iExpression bone,
U (s
i) the interior space of expression skeleton cylinder.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2007100442191A CN100543775C (en) | 2007-07-26 | 2007-07-26 | The method of following the tracks of based on the 3 d human motion of many orders camera |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2007100442191A CN100543775C (en) | 2007-07-26 | 2007-07-26 | The method of following the tracks of based on the 3 d human motion of many orders camera |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101154289A true CN101154289A (en) | 2008-04-02 |
| CN100543775C CN100543775C (en) | 2009-09-23 |
Family
ID=39255932
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2007100442191A Expired - Fee Related CN100543775C (en) | 2007-07-26 | 2007-07-26 | The method of following the tracks of based on the 3 d human motion of many orders camera |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100543775C (en) |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102074034A (en) * | 2011-01-06 | 2011-05-25 | 西安电子科技大学 | Multi-model human motion tracking method |
| CN102279979A (en) * | 2010-06-12 | 2011-12-14 | 陈姝 | Method for calculating scale factor in perspective projection imaging model by utilizing bone constraints |
| CN101692284B (en) * | 2009-07-24 | 2012-01-04 | 西安电子科技大学 | Three-dimensional human body motion tracking method based on quantum immune clone algorithm |
| CN102411783A (en) * | 2010-10-14 | 2012-04-11 | 微软公司 | Automatically tracking user movement in a video chat application |
| CN101789126B (en) * | 2010-01-26 | 2012-12-26 | 北京航空航天大学 | Three-dimensional human body motion tracking method based on volume pixels |
| CN101789125B (en) * | 2010-01-26 | 2013-10-30 | 北京航空航天大学 | Method for tracking human skeleton motion in unmarked monocular video |
| CN103871106A (en) * | 2012-12-14 | 2014-06-18 | 韩国电子通信研究院 | Method of fitting virtual item using human body model and system for providing fitting service of virtual item |
| CN104364823A (en) * | 2012-06-14 | 2015-02-18 | 高通股份有限公司 | Adaptive switching between a vision aided inertial camera pose estimation and a vision based only camera pose estimation |
| CN104700433A (en) * | 2015-03-24 | 2015-06-10 | 中国人民解放军国防科学技术大学 | Vision-based real-time general movement capturing method and system for human body |
| CN106888369A (en) * | 2015-01-13 | 2017-06-23 | 苏州创捷传媒展览股份有限公司 | Virtual telescope interactive device |
| CN107341844A (en) * | 2017-06-21 | 2017-11-10 | 上海大学 | A kind of real-time three-dimensional people's object plotting method based on more Kinect |
| CN107403222A (en) * | 2017-07-19 | 2017-11-28 | 燕山大学 | A kind of motion tracking method based on auxiliary more new model and validity check |
| CN108765577A (en) * | 2018-04-09 | 2018-11-06 | 华南农业大学 | A kind of four limbs farming animals skeleton augmented reality tracking of real-time point cloud data driving |
| CN110675474A (en) * | 2019-08-16 | 2020-01-10 | 咪咕动漫有限公司 | Virtual character model learning method, electronic device and readable storage medium |
| CN110827196A (en) * | 2018-09-05 | 2020-02-21 | 天目爱视（北京）科技有限公司 | Device capable of simultaneously acquiring 3D information of multiple regions of target object |
| CN111028271A (en) * | 2019-12-06 | 2020-04-17 | 浩云科技股份有限公司 | Multi-camera personnel three-dimensional positioning and tracking system based on human skeleton detection |
| CN111062987A (en) * | 2018-09-05 | 2020-04-24 | 天目爱视（北京）科技有限公司 | Virtual matrix type three-dimensional measurement and information acquisition device based on multiple acquisition regions |
| CN111080712A (en) * | 2019-12-06 | 2020-04-28 | 浩云科技股份有限公司 | Multi-camera personnel positioning, tracking and displaying method based on human body skeleton detection |
| CN112254679A (en) * | 2020-10-15 | 2021-01-22 | 天目爱视（北京）科技有限公司 | Multi-position combined 3D acquisition system and method |
| CN112785680A (en) * | 2019-11-07 | 2021-05-11 | 上海莉莉丝科技股份有限公司 | Method, system, device and medium for describing object relationship in three-dimensional virtual space |
| CN113658319A (en) * | 2021-05-17 | 2021-11-16 | 海南师范大学 | Method and device for gesture migration between heterogeneous frameworks |
| WO2022078433A1 (en) * | 2020-10-15 | 2022-04-21 | 左忠斌 | Multi-location combined 3d image acquisition system and method |
-
2007
- 2007-07-26 CN CNB2007100442191A patent/CN100543775C/en not_active Expired - Fee Related
Cited By (33)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101692284B (en) * | 2009-07-24 | 2012-01-04 | 西安电子科技大学 | Three-dimensional human body motion tracking method based on quantum immune clone algorithm |
| CN101789126B (en) * | 2010-01-26 | 2012-12-26 | 北京航空航天大学 | Three-dimensional human body motion tracking method based on volume pixels |
| CN101789125B (en) * | 2010-01-26 | 2013-10-30 | 北京航空航天大学 | Method for tracking human skeleton motion in unmarked monocular video |
| CN102279979A (en) * | 2010-06-12 | 2011-12-14 | 陈姝 | Method for calculating scale factor in perspective projection imaging model by utilizing bone constraints |
| CN102411783B (en) * | 2010-10-14 | 2016-07-06 | 微软技术许可有限责任公司 | Move from motion tracking user in Video chat is applied |
| CN102411783A (en) * | 2010-10-14 | 2012-04-11 | 微软公司 | Automatically tracking user movement in a video chat application |
| US9628755B2 (en) | 2010-10-14 | 2017-04-18 | Microsoft Technology Licensing, Llc | Automatically tracking user movement in a video chat application |
| CN102074034B (en) * | 2011-01-06 | 2013-11-06 | 西安电子科技大学 | Multi-model human motion tracking method |
| CN102074034A (en) * | 2011-01-06 | 2011-05-25 | 西安电子科技大学 | Multi-model human motion tracking method |
| CN104364823A (en) * | 2012-06-14 | 2015-02-18 | 高通股份有限公司 | Adaptive switching between a vision aided inertial camera pose estimation and a vision based only camera pose estimation |
| CN104364823B (en) * | 2012-06-14 | 2017-07-18 | 高通股份有限公司 | Adaptivity switching between vision supplementary inertial camera pose estimation and the camera pose estimation for being based only upon vision |
| CN103871106A (en) * | 2012-12-14 | 2014-06-18 | 韩国电子通信研究院 | Method of fitting virtual item using human body model and system for providing fitting service of virtual item |
| CN106888369A (en) * | 2015-01-13 | 2017-06-23 | 苏州创捷传媒展览股份有限公司 | Virtual telescope interactive device |
| CN104700433A (en) * | 2015-03-24 | 2015-06-10 | 中国人民解放军国防科学技术大学 | Vision-based real-time general movement capturing method and system for human body |
| CN107341844A (en) * | 2017-06-21 | 2017-11-10 | 上海大学 | A kind of real-time three-dimensional people's object plotting method based on more Kinect |
| CN107403222A (en) * | 2017-07-19 | 2017-11-28 | 燕山大学 | A kind of motion tracking method based on auxiliary more new model and validity check |
| CN108765577A (en) * | 2018-04-09 | 2018-11-06 | 华南农业大学 | A kind of four limbs farming animals skeleton augmented reality tracking of real-time point cloud data driving |
| CN108765577B (en) * | 2018-04-09 | 2021-07-09 | 华南农业大学 | Real-time point cloud data-driven four-limb livestock animal skeleton augmented reality tracking method |
| CN110827196A (en) * | 2018-09-05 | 2020-02-21 | 天目爱视（北京）科技有限公司 | Device capable of simultaneously acquiring 3D information of multiple regions of target object |
| CN111062987A (en) * | 2018-09-05 | 2020-04-24 | 天目爱视（北京）科技有限公司 | Virtual matrix type three-dimensional measurement and information acquisition device based on multiple acquisition regions |
| CN110675474B (en) * | 2019-08-16 | 2023-05-02 | 咪咕动漫有限公司 | Learning method for virtual character model, electronic device, and readable storage medium |
| CN110675474A (en) * | 2019-08-16 | 2020-01-10 | 咪咕动漫有限公司 | Virtual character model learning method, electronic device and readable storage medium |
| CN112785680B (en) * | 2019-11-07 | 2023-01-24 | 上海莉莉丝科技股份有限公司 | Method, system, device and medium for describing object relationship in three-dimensional virtual space |
| CN112785680A (en) * | 2019-11-07 | 2021-05-11 | 上海莉莉丝科技股份有限公司 | Method, system, device and medium for describing object relationship in three-dimensional virtual space |
| CN111080712B (en) * | 2019-12-06 | 2023-04-18 | 浩云科技股份有限公司 | Multi-camera personnel positioning, tracking and displaying method based on human body skeleton detection |
| CN111028271B (en) * | 2019-12-06 | 2023-04-14 | 浩云科技股份有限公司 | Multi-camera personnel three-dimensional positioning and tracking system based on human skeleton detection |
| CN111080712A (en) * | 2019-12-06 | 2020-04-28 | 浩云科技股份有限公司 | Multi-camera personnel positioning, tracking and displaying method based on human body skeleton detection |
| CN111028271A (en) * | 2019-12-06 | 2020-04-17 | 浩云科技股份有限公司 | Multi-camera personnel three-dimensional positioning and tracking system based on human skeleton detection |
| WO2022078433A1 (en) * | 2020-10-15 | 2022-04-21 | 左忠斌 | Multi-location combined 3d image acquisition system and method |
| CN112254679A (en) * | 2020-10-15 | 2021-01-22 | 天目爱视（北京）科技有限公司 | Multi-position combined 3D acquisition system and method |
| CN112254679B (en) * | 2020-10-15 | 2023-04-28 | 天目爱视（北京）科技有限公司 | Multi-position combined type 3D acquisition system and method |
| CN113658319A (en) * | 2021-05-17 | 2021-11-16 | 海南师范大学 | Method and device for gesture migration between heterogeneous frameworks |
| CN113658319B (en) * | 2021-05-17 | 2023-08-04 | 海南师范大学 | Gesture migration method and device between heterogeneous frameworks |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100543775C (en) | 2009-09-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100543775C (en) | The method of following the tracks of based on the 3 d human motion of many orders camera | |
| CN109636831B (en) | Method for estimating three-dimensional human body posture and hand information | |
| CN110458939B (en) | Indoor scene modeling method based on visual angle generation | |
| CN108830150B (en) | One kind being based on 3 D human body Attitude estimation method and device | |
| CN111968129B (en) | Instant positioning and map construction system and method with semantic perception | |
| Yang et al. | S3: Neural shape, skeleton, and skinning fields for 3d human modeling | |
| CN109840940B (en) | Dynamic three-dimensional reconstruction method, device, equipment, medium and system | |
| CN109671120A (en) | A kind of monocular SLAM initial method and system based on wheel type encoder | |
| CN100407798C (en) | Three-dimensional geometric mode building system and method | |
| CN108416840A (en) | A kind of dense method for reconstructing of three-dimensional scenic based on monocular camera | |
| CN102467753B (en) | Method and system for reconstructing time-varying point cloud based on framework registration | |
| CN106384383A (en) | RGB-D and SLAM scene reconfiguration method based on FAST and FREAK feature matching algorithm | |
| CN104036488B (en) | Binocular vision-based human body posture and action research method | |
| CN104915978B (en) | Realistic animation generation method based on body-sensing camera Kinect | |
| CN106780592A (en) | Kinect depth reconstruction algorithms based on camera motion and image light and shade | |
| CN107240129A (en) | Object and indoor small scene based on RGB D camera datas recover and modeling method | |
| CN106127739A (en) | A kind of RGB D SLAM method of combination monocular vision | |
| CN111062326B (en) | Self-supervision human body 3D gesture estimation network training method based on geometric driving | |
| CN111553968A (en) | Method for reconstructing animation by three-dimensional human body | |
| CN108629294A (en) | Human body based on deformation pattern and face net template approximating method | |
| CN113421328B (en) | Three-dimensional human body virtual reconstruction method and device | |
| CN106815855A (en) | Based on the human body motion tracking method that production and discriminate combine | |
| CN110135277B (en) | Human behavior recognition method based on convolutional neural network | |
| CN111797692A (en) | Depth image gesture estimation method based on semi-supervised learning | |
| CN117315169A (en) | Live-action three-dimensional model reconstruction method and system based on deep learning multi-view dense matching |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090923 Termination date: 20120726 |