CN106599873A - Figure identity identification method based on three-dimensional attitude information - Google Patents
Figure identity identification method based on three-dimensional attitude information Download PDFInfo
- Publication number
- CN106599873A CN106599873A CN201611206255.9A CN201611206255A CN106599873A CN 106599873 A CN106599873 A CN 106599873A CN 201611206255 A CN201611206255 A CN 201611206255A CN 106599873 A CN106599873 A CN 106599873A
- Authority
- CN
- China
- Prior art keywords
- information
- dimensional
- pose
- piece identity
- recognition methodss
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/64—Three-dimensional objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Human Computer Interaction (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a figure identity identification method based on three-dimensional attitude information. The figure identity identification method based on three-dimensional attitude information comprises the following step: step one, acquiring an image sequence of monitored scenes, detecting human bodies in each acquired image, and obtaining three-dimensional information of all human bodies, the three-dimensional information concretely including the three-dimensional space positions, heights, shapes, orientations and motion states of the human bodies; step two, tracking each human body on the image frame sequence and calculating the three-dimensional information of each human body; step three, through a cumulative size distribution diagram technology, obtaining attitude information of the human bodies on the basis of three-dimensional sequence information of the tracked human bodies; step four, training classifiers and carrying out real-time classification according to the attitude information obtained in the step three; and step five, identifying the human bodies. The figure identity identification method avoids a problem of a conventional biology identification technology which demands contacting or close-range identification. The invention provides the figure identity identification method that can identify a figure in a long distance or in a condition that the face of the figure is blocked.
Description
Technical field
The present invention relates to 3 d pose identification field, in particular it relates to be based on piece identity's identification of 3 d pose information
Method.
Background technology
In recent years, gesture recognition is increasingly becoming the hot issue of computer vision and field of Computer Graphics, extensively should
The scene such as block for remote piece identity's identification such as safety monitoring, video search and face, at present by gesture recognition come
Identification to piece identity also achieves certain progress, and by the analysis of the two-dimentional gait information to personage personage is confirmed
Identity.But in prior art when recognizing to piece identity, the PTZ shapes that can be subject to camera are solved due to two-dimentional gait
The factor such as the distance of state, the visual angle of camera, camera and target and the direction of motion of target affects, and cannot be in different equipment
Piece identity is accurately determined with scene.
Therefore it provides one kind is in use, not only can be known by the 3 d pose information under cordless
Other piece identity, can be identified at a distance to the identity of personage, and judged result is accurately based on 3 d pose information
Piece identity's recognition methodss are the problems of urgent need to resolve of the present invention.
The content of the invention
For above-mentioned technical problem, the purpose of the present invention is overcome in prior art when recognizing to piece identity, due to
The distance and the motion side of target that solve PTZ states, the visual angle of camera, camera and the target that can be subject to camera of two-dimentional gait
To etc. factor affect, and the problem of piece identity cannot be accurately determined under different equipment and scene, so as to provide one kind
In use, not only piece identity can be recognized by the 3 d pose information under cordless, can be right at a distance
The identity of personage is identified, and judged result piece identity's recognition methodss accurately based on 3 d pose information.
To achieve these goals, the invention provides a kind of piece identity's recognition methodss based on 3 d pose information,
Piece identity's recognition methodss based on 3 d pose information include：Step 1, the image sequence of acquisition monitoring scene is being adopted
Human body is detected in the every two field picture for collecting, and three-dimensional information acquisition is carried out to everyone；Step 2 is right in picture frame sequence
Everyone is tracked and calculates its three-dimensional information；Step 3, using the human body three-dimensional sequence information obtained by tracking everyone is obtained
Attitude information；Step 4, is trained grader and is carried out real-time grading using the attitude information obtained in step 3；Step 5, it is complete
Into the identification of piece identity.
Preferably, the three-dimensional information of the people also includes：The three-dimensional space position of personage, height, build, body direction and
Kinestate.
Preferably, the image sequence that monitoring scene is obtained using the visual system through demarcation is needed in the step 1
With proprietary three-dimensional information.
Preferably, cumulative distribution diagram technology and the human body three-dimensional sequence information with reference to obtained by tracking used in the step 3
To obtain the attitude information of people.
Preferably, it is further comprising the steps of between step 1 and step 2：According to the three-dimensional information image of the personage for obtaining
Estimate out the 3 d pose of personage.
Preferably, using the 3 d pose that personage is estimated based on the method for geometric properties.
According to above-mentioned technical proposal, the piece identity's recognition methodss based on 3 d pose information that the present invention is provided are by obtaining
After taking the image sequence of monitoring scene, human body is detected in described image sequence, and obtain the three-dimensional information of human body, then
Everyone is tracked, so as to draw the three-dimensional series information of human body, obtains everyone by the three-dimensional series information for obtaining
Attitude information, retraining grader carries out real-time grading, so as to draw the identity of personage.The present invention provide based on 3 d pose
Piece identity's recognition methodss of information are overcome in prior art when recognizing to piece identity, because the solution of two-dimentional gait can be received
The factors such as the direction of motion of PTZ states, the visual angle of camera, the distance of camera and target and target to camera affect, and nothing
Method accurately determines the problem of piece identity under different equipment and scene.
Other features and advantages of the present invention will be described in detail in subsequent specific embodiment part.
Description of the drawings
Accompanying drawing is, for providing a further understanding of the present invention, and to constitute the part of description, with following tool
Body embodiment is used to explain the present invention together, but is not construed as limiting the invention.In the accompanying drawings：
Fig. 1 is a kind of stream of piece identity's recognition methodss that the present invention is based on 3 d pose information under preferred implementation
Cheng Tu；
Fig. 2 is that the present invention is based on prison in piece identity's recognition methodss of 3 d pose information under a kind of preferred implementation
The image effect figure of single human body in control scene；
Fig. 3 is that the present invention is based on list in piece identity's recognition methodss of 3 d pose information under a kind of preferred implementation
The cumulative distribution design sketch of individual human body；
Fig. 4 is that of the invention being based in piece identity's recognition methodss of 3 d pose information under a kind of preferred implementation is tired out
The three-dimensional image effect figure of integration Butut.
Specific embodiment
The specific embodiment of the present invention is described in detail below in conjunction with accompanying drawing.It should be appreciated that this place is retouched
The specific embodiment stated is merely to illustrate and explains the present invention, is not limited to the present invention.
Such as Fig. 1, the invention provides a kind of piece identity's recognition methodss based on 3 d pose information, described based on three-dimensional
Piece identity's recognition methodss of attitude information include：Step 1, the image sequence of acquisition monitoring scene, in the every frame figure for collecting
Human body is detected as in, and three-dimensional information acquisition is carried out to everyone；Step 2, is tracked in picture frame sequence to everyone
And calculate its three-dimensional information；Step 3, using the human body three-dimensional sequence information obtained by tracking proprietary attitude information is obtained；Step
Rapid 4, train grader and carry out real-time grading using the attitude information obtained in step 3；Step 5, completes piece identity's
Identification.
According to above-mentioned technical proposal, the piece identity's recognition methodss based on 3 d pose information that the present invention is provided are by obtaining
After taking the image sequence of monitoring scene, human body is detected in described image sequence, and obtain the three-dimensional information of human body, then
Everyone is tracked, so as to draw the three-dimensional series information of human body, is obtained by the three-dimensional series information for obtaining proprietary
Attitude information, retraining grader carries out real-time grading, so as to draw the identity of personage.The present invention provide based on 3 d pose
Piece identity's recognition methodss of information are overcome in prior art when recognizing to piece identity, due to the solution meeting of two-dimentional gait
The factors such as the direction of motion of PTZ states, the visual angle of camera, the distance of camera and target and target by camera are affected, and
The problem of piece identity cannot be accurately determined under different equipment and scene.
The present invention it is a kind of preferred embodiment in, in order that the attitude information of the people for obtaining is more accurate,
Convenient detection and analysis calculates its three-dimensional information, and the three-dimensional information of the people also includes：The three-dimensional space position of personage, height, body
Type, body direction and kinestate, by these three-dimensional informations the information of human body can be effectively grasped, convenient accurately to obtain
The attitude information of people, in numerous information of human body, the information that most can at a distance recognize piece identity is the gait information of people,
The three-dimensional information of people is obtained in step 1 also including the gait information of people, can more fast and accurately be connected by gait information
Oneself information of personage, conveniently obtains the attitude information of people.
In the present invention, in order to obtain the picture frame sequence of target person more fully hereinafter, consequently facilitating detect human body and
Three-dimensional information, the present invention it is a kind of preferred embodiment in, need to carry out background to monitoring scene in the step 1 to build
Mould.
The present invention it is a kind of preferred embodiment in, need in the step 1 using through the total space demarcate regarding
Feel system is obtaining the image sequence and proprietary three-dimensional information of monitoring scene.
In the present invention, in order that the attitude information of the people drawn in the step 3 is more accurately, the one of the present invention
In planting preferred embodiment, the present invention is using cumulative distribution diagram technology and combines the human body three-dimensional sequence information obtained by tracking
Obtain the attitude information of people.
The present invention it is a kind of preferred embodiment in, it is further comprising the steps of between step 1 and step 2：According to
The picture frame sequence of the target person of acquisition estimates out the 3 d pose of personage, and the three-dimensional appearance of personage is accurately obtained for convenience
State, after the picture frame sequence for obtaining target person, can estimate out the 3 d pose of personage according to image frame sequence column information,
So can also increase the accuracy and seriality of the attitude information for obtaining people.
In the present invention, when the 3 d pose to personage is estimated, can adopt based on the method for statistical classification, but
Be the present invention it is a kind of preferred embodiment in, using the 3 d pose that personage is estimated based on the method for geometric properties,
The so convenient 3 d pose for drawing personage, and accuracy is high.
The preferred embodiment of the present invention is described in detail above in association with accompanying drawing, but, the present invention is not limited to above-mentioned reality
The detail in mode is applied, in the range of the technology design of the present invention, various letters can be carried out to technical scheme
Monotropic type, these simple variants belong to protection scope of the present invention.
It is further to note that each particular technique feature described in above-mentioned specific embodiment, in not lance
In the case of shield, can be combined by any suitable means, in order to avoid unnecessary repetition, the present invention to it is various can
The compound mode of energy is no longer separately illustrated.
Additionally, combination in any can also be carried out between a variety of embodiments of the present invention, as long as it is without prejudice to this
The thought of invention, it should equally be considered as content disclosed in this invention.
Claims (6)
1. a kind of piece identity's recognition methodss based on 3 d pose information, it is characterised in that described based on 3 d pose information
Piece identity's recognition methodss include：
Step 1, the image sequence of acquisition monitoring scene detects human body in the every two field picture for collecting, and everyone is entered
Row three-dimensional information is obtained；
Step 2, is tracked and calculates its three-dimensional information in picture frame sequence to everyone；
Step 3, using the human body three-dimensional sequence information obtained by tracking everyone attitude information is obtained；
Step 4, is trained grader and is carried out real-time grading using the attitude information obtained in step 4；
Step 5, completes the identification of piece identity.
2. the piece identity's recognition methodss based on 3 d pose information according to claim 1, it is characterised in that step 1
Position, height, build, body direction and kinestate that the middle three-dimensional information for obtaining people is behaved.
3. the piece identity's recognition methodss based on 3 d pose information according to claim 1, it is characterised in that the step
The image sequence and proprietary three-dimensional information that monitoring scene is obtained using the visual system through demarcation is needed in rapid 1.
4. the piece identity's recognition methodss based on 3 d pose information according to claim 1, it is characterised in that the step
Used in rapid 3 cumulative distribution diagram technology and combine tracking obtained by human body three-dimensional sequence information come obtain everyone attitude believe
Breath.
5. the piece identity's recognition methodss based on 3 d pose information according to claim 1, it is characterised in that in step
It is further comprising the steps of between 1 and step 2：The 3 d pose of personage is estimated out according to the three-dimensional information image of the personage for obtaining.
6. the piece identity's recognition methodss based on 3 d pose information according to claim 5, it is characterised in that use base
The 3 d pose of personage is estimated in the method for geometric properties.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611206255.9A CN106599873A (en) | 2016-12-23 | 2016-12-23 | Figure identity identification method based on three-dimensional attitude information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611206255.9A CN106599873A (en) | 2016-12-23 | 2016-12-23 | Figure identity identification method based on three-dimensional attitude information |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106599873A true CN106599873A (en) | 2017-04-26 |
Family
ID=58603465
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611206255.9A Pending CN106599873A (en) | 2016-12-23 | 2016-12-23 | Figure identity identification method based on three-dimensional attitude information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106599873A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107292252A (en) * | 2017-06-09 | 2017-10-24 | 南京华捷艾米软件科技有限公司 | A kind of personal identification method of autonomous learning |
| CN108564017A (en) * | 2018-04-04 | 2018-09-21 | 北京天目智联科技有限公司 | A kind of biological characteristic 3D 4 D datas recognition methods and system based on grating camera |
| CN110297929A (en) * | 2019-06-14 | 2019-10-01 | 北京达佳互联信息技术有限公司 | Image matching method, device, electronic equipment and storage medium |
| CN112800885A (en) * | 2021-01-16 | 2021-05-14 | 南京众鑫云创软件科技有限公司 | Data processing system and method based on big data |
| CN113140051A (en) * | 2020-01-20 | 2021-07-20 | 上海依图信息技术有限公司 | Attendance checking method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101558996A (en) * | 2009-05-15 | 2009-10-21 | 天津大学 | Gait recognition method based on orthogonal projection three-dimensional reconstruction of human motion structure |
| CN102697508A (en) * | 2012-04-23 | 2012-10-03 | 中国人民解放军国防科学技术大学 | Method for performing gait recognition by adopting three-dimensional reconstruction of monocular vision |
| CN103177269A (en) * | 2011-12-23 | 2013-06-26 | 北京三星通信技术研究有限公司 | Equipment and method used for estimating object posture |
| CN103729614A (en) * | 2012-10-16 | 2014-04-16 | 上海唐里信息技术有限公司 | People recognition method and device based on video images |
| US20160042227A1 (en) * | 2014-08-06 | 2016-02-11 | BAE Systems Information and Electronic Systems Integraton Inc. | System and method for determining view invariant spatial-temporal descriptors for motion detection and analysis |
-
2016
- 2016-12-23 CN CN201611206255.9A patent/CN106599873A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101558996A (en) * | 2009-05-15 | 2009-10-21 | 天津大学 | Gait recognition method based on orthogonal projection three-dimensional reconstruction of human motion structure |
| CN103177269A (en) * | 2011-12-23 | 2013-06-26 | 北京三星通信技术研究有限公司 | Equipment and method used for estimating object posture |
| CN102697508A (en) * | 2012-04-23 | 2012-10-03 | 中国人民解放军国防科学技术大学 | Method for performing gait recognition by adopting three-dimensional reconstruction of monocular vision |
| CN103729614A (en) * | 2012-10-16 | 2014-04-16 | 上海唐里信息技术有限公司 | People recognition method and device based on video images |
| US20160042227A1 (en) * | 2014-08-06 | 2016-02-11 | BAE Systems Information and Electronic Systems Integraton Inc. | System and method for determining view invariant spatial-temporal descriptors for motion detection and analysis |
Non-Patent Citations (1)
| Title |
|---|
| 贾立好 等: "基于头顶点三维运动轨迹的身份识别新方法", 《自动化学报》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107292252A (en) * | 2017-06-09 | 2017-10-24 | 南京华捷艾米软件科技有限公司 | A kind of personal identification method of autonomous learning |
| CN108564017A (en) * | 2018-04-04 | 2018-09-21 | 北京天目智联科技有限公司 | A kind of biological characteristic 3D 4 D datas recognition methods and system based on grating camera |
| CN110297929A (en) * | 2019-06-14 | 2019-10-01 | 北京达佳互联信息技术有限公司 | Image matching method, device, electronic equipment and storage medium |
| CN113140051A (en) * | 2020-01-20 | 2021-07-20 | 上海依图信息技术有限公司 | Attendance checking method and device, electronic equipment and storage medium |
| CN112800885A (en) * | 2021-01-16 | 2021-05-14 | 南京众鑫云创软件科技有限公司 | Data processing system and method based on big data |
| CN112800885B (en) * | 2021-01-16 | 2023-09-26 | 南京众鑫云创软件科技有限公司 | Data processing system and method based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105930767B (en) | A kind of action identification method based on human skeleton | |
| Droeschel et al. | Learning to interpret pointing gestures with a time-of-flight camera | |
| CN106599873A (en) | Figure identity identification method based on three-dimensional attitude information | |
| CN103824070B (en) | A kind of rapid pedestrian detection method based on computer vision | |
| CN105260705B (en) | A kind of driver's making and receiving calls behavioral value method suitable under multi-pose | |
| CN110837784A (en) | Examination room peeping cheating detection system based on human head characteristics | |
| CN102156537A (en) | Equipment and method for detecting head posture | |
| Sáez et al. | Aerial obstacle detection with 3-D mobile devices | |
| Droeschel et al. | Towards joint attention for a domestic service robot-person awareness and gesture recognition using time-of-flight cameras | |
| Denman et al. | Soft-biometrics: unconstrained authentication in a surveillance environment | |
| CN105868694A (en) | Dual-mode emotion identification method and system based on facial expression and eyeball movement | |
| CN114067358A (en) | Human body posture recognition method and system based on key point detection technology | |
| US20120087543A1 (en) | Image-based hand detection apparatus and method | |
| CN104700088A (en) | Gesture track recognition method based on monocular vision motion shooting | |
| Appenrodt et al. | Data gathering for gesture recognition systems based on single color-, stereo color-and thermal cameras | |
| CN102831408A (en) | Human face recognition method | |
| EP2629241A1 (en) | Control of a wearable device | |
| CN109308718A (en) | A kind of space personnel positioning apparatus and method based on more depth cameras | |
| CN104821010A (en) | Binocular-vision-based real-time extraction method and system for three-dimensional hand information | |
| CN103995595A (en) | Game somatosensory control method based on hand gestures | |
| CN104915656A (en) | Quick human face recognition method based on binocular vision measurement technology | |
| KR20170082074A (en) | Face recognition apparatus and method using physiognomic information | |
| CN109934127A (en) | Pedestrian's recognition and tracking method based on video image and wireless signal | |
| Li et al. | Robust multiperson detection and tracking for mobile service and social robots | |
| CN106650628B (en) | Fingertip detection method based on three-dimensional K curvature |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170426 |