CN103714555B - A kind of four-dimensional motion point cloud segmentation and method for reconstructing based on movement locus - Google Patents
A kind of four-dimensional motion point cloud segmentation and method for reconstructing based on movement locus Download PDFInfo
- Publication number
- CN103714555B CN103714555B CN201310684707.4A CN201310684707A CN103714555B CN 103714555 B CN103714555 B CN 103714555B CN 201310684707 A CN201310684707 A CN 201310684707A CN 103714555 B CN103714555 B CN 103714555B
- Authority
- CN
- China
- Prior art keywords
- data
- movement locus
- frame
- cluster analysis
- motion trace
- 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.)
- Active
Links
- 230000033001 locomotion Effects 0.000 title claims abstract description 69
- 238000000034 method Methods 0.000 title claims abstract description 44
- 230000011218 segmentation Effects 0.000 title claims abstract description 15
- 238000007621 cluster analysis Methods 0.000 claims abstract description 24
- 238000000605 extraction Methods 0.000 claims abstract description 17
- 238000013075 data extraction Methods 0.000 claims description 6
- 230000003595 spectral effect Effects 0.000 claims description 2
- 238000010606 normalization Methods 0.000 claims 2
- 238000005457 optimization Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000011524 similarity measure Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
Landscapes
- Length Measuring Devices With Unspecified Measuring Means (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The present invention relates to area of computer graphics, there is provided a kind of four-dimensional motion point cloud segmentation and method for reconstructing based on movement locus.The skeletal extraction of movement locus is realized, including：Exercise data is obtained, adjacent data registration is carried out, motion trace data is extracted, and motion trace data cluster analysis, the exercise data after cluster analysis carries out uniformity skeletal extraction.The present invention does not need predefined masterplate, it is not required that the data of each frame carry out skeletal extraction first, relatively low to a cloud quality requirement.
Description
Technical field
The present invention relates to area of computer graphics, more particularly to a kind of four-dimensional motion point cloud segmentation based on movement locus
With method for reconstructing.
Background technology
It is existing it is motion-captured need masterplate or higher to a cloud quality requirement with segmentation, or from single frames point Yun Zhongti
Skeleton is taken out, the uniformity optimization of multiframe skeleton is carried out and is processed.
It is, based on point cloud geometric properties, to lack the continuity in time and space that segmentation of the prior art to object is most of
And uniformity；The existing point cloud framework method for reconstructing for pursuing space-time consistency, the quality requirement to input point cloud is higher, because
They need to extract coarse skeleton from the sequential point cloud of input, then carry out the uniformity optimization in time and space.
The present invention need not in advance extract skeleton, and the uniformity of skeleton is made in consideration during extraction.
The content of the invention
The present invention uses a kind of four-dimensional motion point cloud segmentation and method for reconstructing based on movement locus, realizes movement locus
Skeletal extraction, the present invention do not need predefined masterplate, it is not required that the data of each frame carry out skeletal extraction first, to point
Cloud quality requirement is relatively low.The present invention uses following scheme：
A kind of four-dimensional motion point cloud segmentation and method for reconstructing based on movement locus, including：
S1, acquisition exercise data；
S2, consecutive frame Registration of Measuring Data is carried out to the exercise data；
S3, to by step S2 registration after data carry out motion trace data extraction；
S4, cluster analysis is carried out to the motion trace data described in step S3；
S5, the data to being obtained after cluster analysis described in step S4 carry out uniformity skeletal extraction.
Preferably, the acquisition exercise data, using the exercise data catching method based on laser scanner,
Preferably, laser scanner carries out cloud data scanning and acquisition so that certain frame per second is continuous to moving object；
During data after acquisition are stored in computer in the form of one file of every frame.
Preferably, it is using the method reality of non-rigid matching to the method that the exercise data carries out consecutive frame Registration of Measuring Data
Existing, the non-rigid matching process is that each point in each frame finds a corresponding points in the next frame.
Preferably, it is to the method that motion trace data extraction is carried out by the data after step S2 registrations, using depth
Movement locus is connected growth between adjacent two frame that be similar in adjacent direction by preferential method, so as to obtain between multiframe point with
Movement locus between point.
Preferably, it is to the method that the motion trace data described in step S3 carries out cluster analysis, based on movement locus
Range formula, using spectral clustering or kmean clustering procedures, the similitude according to motion trace data is by moving meshes for not
Same motion parts；
Preferably, it is characterised in that the data to being obtained after cluster analysis described in step S4 carry out uniformity skeletal extraction
Also include the step of part completed to segmentation calculates the skeleton of uniformity according to syntople.
Preferably, the method that the skeleton of uniformity is calculated according to syntople is, to the node of skeleton, from adjacent node
It is next so as to the key node of skeleton be transformed into and adjacent node is to corresponding relation is found in the corresponding points of next frame
Frame, iteration is carried out, in may switch to N frame data.
A kind of four-dimensional motion point cloud segmentation and method for reconstructing based on movement locus disclosed by the invention, are moved by obtaining
Data, carry out adjacent data registration, and motion trace data is extracted, motion trace data cluster analysis, the motion after cluster analysis
Data carry out uniformity skeletal extraction, realize the skeletal extraction of movement locus, and the present invention does not need predefined masterplate, not yet
Need the data of each frame carries out skeletal extraction first, relatively low to a cloud quality requirement.
Brief description of the drawings
Fig. 1 is a kind of four-dimensional motion point cloud segmentation based on movement locus of the embodiment of the present invention 1 and method for reconstructing flow chart；
Fig. 2 be the frame point cloud of the embodiment of the present invention 1 four as an example；
Fig. 3 is the example after the consecutive frame Registration of Measuring Data of the embodiment of the present invention 1；
Fig. 4 be 1 point of the embodiment of the present invention be aligned after point line relation；
Fig. 5 is the movement locus between the multiframe of the embodiment of the present invention 1 between points；
Fig. 6 intuitively shows figure for the distance definition of the different tracks of the embodiment of the present invention 1；
Fig. 7 is some movement locus before the cluster analysis of the embodiment of the present invention 1；
Fig. 8 is the movement locus after the cluster analysis of the embodiment of the present invention 1.
Specific embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
A kind of four-dimensional motion point cloud segmentation based on movement locus and method for reconstructing are the embodiment of the invention provides, including：
S1, acquisition exercise data；
S2, consecutive frame Registration of Measuring Data is carried out to the exercise data；
S3, to by step S2 registration after data carry out motion trace data extraction；
S4, cluster analysis is carried out to the motion trace data described in step S3.
S5, the data to being obtained after cluster analysis described in step S4 carry out uniformity skeletal extraction.
The embodiment of the present invention carries out adjacent data registration by obtaining exercise data, and motion trace data is extracted, and moves rail
Mark data clusters are analyzed, and the exercise data after cluster analysis carries out uniformity skeletal extraction, and the skeleton for realizing movement locus is carried
Take, the present invention does not need predefined masterplate, it is not required that the data of each frame carry out skeletal extraction first, will to a cloud quality
Ask relatively low.The present invention will be described in detail below.
Embodiment 1：
Refer to shown in Fig. 1, be a kind of four-dimensional motion point cloud segmentation based on movement locus of the invention and method for reconstructing stream
Cheng Tu.The method comprises the steps：
S1, acquisition exercise data；
The present embodiment provides a four frame point clouds as an example, such as Fig. 2, is caught using the exercise data based on laser scanner
Method is obtained, laser scanner carries out cloud data scanning and acquisition so that certain frame per second is continuous to moving object；After acquisition
During data are stored in computer in the form of one file of every frame.
S2, consecutive frame Registration of Measuring Data is carried out to the exercise data；
Consecutive frame Registration of Measuring Data is carried out to exercise data, is realized using the method for non-rigid matching, matching process is each
Each point in frame finds a corresponding points in the next frame, such as Fig. 3, is the point after registration per right-hand point in frame, point with it is right
Line relation such as Fig. 4 of point after standard.
S3, to by step S2 registration after data carry out motion trace data extraction；
To carrying out motion trace data extraction by the data after step S2 registrations, will be adjacent using the method for depth-first
Similar adjacent two frame in direction between movement locus be connected growth, so as to obtain the motion rail between multiframe between points
Mark, such as Fig. 5.
S4, cluster analysis is carried out to the motion trace data described in step S3；
Cluster analysis is carried out to the motion trace data described in step S3, the range formula based on movement locus, using spectrum
Cluster or kmean clustering procedures, moving meshes are different motion parts by the similitude according to motion trace data；
Similarity measures between motion trace data use Euclidean distance algorithm, it is considered to Euclidean distance between track and
Relation between the direction of motion.During cluster analysis, N bars track is randomly selected as cluster centre, then by similar track number
According to being constantly added in the middle of nearest cluster, while updating cluster centre, this step iteration proceeds to all of track and is gathered
Class is lived untill meeting the other conditions that user specifies.The definition of different distance is calculated to be made with the following method：
D1=Euclidean distances；
D2=angle αs；
D=normalizes (d1), normalizes (d2)
The intuitively displaying such as Fig. 6 of the distance definition of different tracks.
Some movement locus such as Fig. 7 before cluster analysis, such as movement locus after cluster analysis, Fig. 8
S5, the data to being obtained after cluster analysis described in step S4 carry out uniformity skeletal extraction；
The skeleton of uniformity is calculated according to syntople, including to the node of skeleton, from adjacent node and adjacent
To corresponding relation is found in the corresponding points of next frame, so as to the key node of skeleton is transformed into next frame, iteration is carried out node,
In may switch to N frame data.
Presently preferred embodiments of the present invention is the foregoing is only, is not intended to limit the invention, it is all in essence of the invention
Any modification, equivalent and improvement made within god and principle etc., should be included within the scope of the present invention.
Claims (7)
1. a kind of four-dimensional motion point cloud segmentation and method for reconstructing based on movement locus, it is characterised in that including：
S1, acquisition exercise data；
S2, consecutive frame Registration of Measuring Data is carried out to the exercise data；
S3, to by step S2 registration after data carry out motion trace data extraction；
S4, cluster analysis is carried out to the motion trace data described in step S3；
S5, the data to being obtained after cluster analysis described in step S4 carry out uniformity skeletal extraction；
It is that the range formula based on movement locus is adopted to the method that the motion trace data described in step S3 carries out cluster analysis
By moving meshes it is different motion parts according to motion trace data with spectral clustering or kmean clustering procedures；
Calculated according to motion trace data and use Euclidean distance algorithm, during cluster analysis, randomly select N bars track as in cluster
, then be constantly added to track data in the middle of nearest cluster by the heart, while updating cluster centre, this step iteration proceeds to
Untill all of track is clustered, the definition of different distance is calculated and made with the following method：
D1=Euclidean distances；
D2=angle αs；
D=normalization (d1)+normalization (d2).
2. method according to claim 1, it is characterised in that the acquisition exercise data, using based on laser scanner
Exercise data catching method.
3. method according to claim 2, it is characterised in that laser scanner continuously carries out a cloud number to moving object
According to scanning and acquisition；During data after acquisition are stored in computer in the form of one file of every frame.
4. method according to claim 1, it is characterised in that the side of consecutive frame Registration of Measuring Data is carried out to the exercise data
Method is to be realized using the method for non-rigid matching, the non-rigid matching process be each point in each frame in the next frame
Find a corresponding points.
5. method according to claim 1, it is characterised in that to carrying out movement locus by the data after step S2 registrations
The method that data are extracted is that movement locus is connected and grows between adjacent two frame using the method for depth-first by adjacent direction,
So as to obtain the movement locus between multiframe between points.
6. method according to claim 1, it is characterised in that the data to being obtained after cluster analysis described in step S4 are carried out
Uniformity skeletal extraction also includes the step of part completed to segmentation calculates the skeleton of uniformity according to syntople.
7. method according to claim 6, it is characterised in that the method that the skeleton of uniformity is calculated according to syntople
To the node of skeleton, corresponding relation to be found from adjacent node and adjacent node to the corresponding points of next frame, so that will
The key node of skeleton is transformed into next frame, and iteration is carried out, in may switch to N frame data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310684707.4A CN103714555B (en) | 2013-12-13 | 2013-12-13 | A kind of four-dimensional motion point cloud segmentation and method for reconstructing based on movement locus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310684707.4A CN103714555B (en) | 2013-12-13 | 2013-12-13 | A kind of four-dimensional motion point cloud segmentation and method for reconstructing based on movement locus |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103714555A CN103714555A (en) | 2014-04-09 |
| CN103714555B true CN103714555B (en) | 2017-06-13 |
Family
ID=50407498
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310684707.4A Active CN103714555B (en) | 2013-12-13 | 2013-12-13 | A kind of four-dimensional motion point cloud segmentation and method for reconstructing based on movement locus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103714555B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104504897B (en) * | 2014-09-28 | 2017-10-31 | 北京工业大学 | A kind of analysis of intersection traffic properties of flow and vehicle movement Forecasting Methodology based on track data |
| CN105006007B (en) * | 2015-07-23 | 2018-06-29 | 北京理工大学 | A kind of action method for reconstructing of data-driven |
| CN105912983B (en) * | 2016-04-01 | 2019-05-24 | 北京理工大学 | Movement method for reconstructing based on sparse input signal |
| CN106204635B (en) * | 2016-06-27 | 2018-11-30 | 北京工业大学 | Based on L0The human body successive frame bone optimization method of minimum |
| CN110378904B (en) * | 2018-07-09 | 2021-10-01 | 北京京东尚科信息技术有限公司 | Method and device for segmenting point cloud data |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102467753A (en) * | 2010-11-04 | 2012-05-23 | 中国科学院深圳先进技术研究院 | Method and system for reconstructing time-varying point cloud based on framework registration |
| CN103218817A (en) * | 2013-04-19 | 2013-07-24 | 深圳先进技术研究院 | Partition method and partition system of plant organ point clouds |
| CN103268631A (en) * | 2013-05-23 | 2013-08-28 | 中国科学院深圳先进技术研究院 | Method and device for extracting point cloud framework |
-
2013
- 2013-12-13 CN CN201310684707.4A patent/CN103714555B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102467753A (en) * | 2010-11-04 | 2012-05-23 | 中国科学院深圳先进技术研究院 | Method and system for reconstructing time-varying point cloud based on framework registration |
| CN103218817A (en) * | 2013-04-19 | 2013-07-24 | 深圳先进技术研究院 | Partition method and partition system of plant organ point clouds |
| CN103268631A (en) * | 2013-05-23 | 2013-08-28 | 中国科学院深圳先进技术研究院 | Method and device for extracting point cloud framework |
Non-Patent Citations (1)
| Title |
|---|
| 《基于结构相似度的轨迹聚类算法》;袁冠等;《通 信 学 报》;20110930;第32卷(第9期);103-110 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103714555A (en) | 2014-04-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103714555B (en) | A kind of four-dimensional motion point cloud segmentation and method for reconstructing based on movement locus | |
| CN109614922B (en) | Dynamic and static gesture recognition method and system | |
| CN105550699B (en) | A kind of video identification classification method based on CNN fusion space-time remarkable information | |
| Zhou et al. | Salient region detection via integrating diffusion-based compactness and local contrast | |
| Lähner et al. | SHREC'16: Matching of deformable shapes with topological noise | |
| CN105940392B (en) | The image-editing technology of device | |
| EP2864930B1 (en) | Self learning face recognition using depth based tracking for database generation and update | |
| US20110286673A1 (en) | Method system and associated modules and software components for providing image sensor based human machine interfacing | |
| CN110032942A (en) | Action identification method based on Time Domain Piecewise and signature differential | |
| US9070043B2 (en) | Method and apparatus for analyzing video based on spatiotemporal patterns | |
| JP2018507476A (en) | Screening for computer vision | |
| CN105493078B (en) | Colored sketches picture search | |
| EP2846309A1 (en) | Method and apparatus for segmenting object in image | |
| CN109614933B (en) | Motion segmentation method based on deterministic fitting | |
| CN103218603A (en) | Face automatic labeling method and system | |
| KR101439037B1 (en) | Method and apparatus for tracking object in image | |
| CN105678810B (en) | Based on the tracking cell method that global and local level is optimal | |
| Liu et al. | Automatic motion capture data denoising via filtered subspace clustering and low rank matrix approximation | |
| CN103886318A (en) | Method for extracting and analyzing nidus areas in pneumoconiosis gross imaging | |
| KR101642200B1 (en) | Apparatus and method for generating motion effects by analyzing motion of object | |
| CN110008841A (en) | A kind of Expression Recognition model building method and system | |
| CN104200441A (en) | Higher-order singular value decomposition based magnetic resonance image denoising method | |
| CN104200218A (en) | Cross-view-angle action identification method and system based on time sequence information | |
| CN108021921A (en) | Image characteristic point extraction system and its application | |
| CN107644233A (en) | FILTERSIM analogy methods based on Cluster Classification |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20140409 Assignee: SHENZHEN YIHUO TECHNOLOGY CO.,LTD. Assignor: SHENZHEN INSTITUTES OF ADVANCED TECHNOLOGY CHINESE ACADEMY OF SCIENCES Contract record no.: X2023980043401 Denomination of invention: A four-dimensional motion point cloud segmentation and reconstruction method based on motion trajectory Granted publication date: 20170613 License type: Common License Record date: 20231013 |
|
| EE01 | Entry into force of recordation of patent licensing contract |