CN104408769A - Virtual netmeeting method based on three-dimensional face recognition - Google Patents
Virtual netmeeting method based on three-dimensional face recognition Download PDFInfo
- Publication number
- CN104408769A CN104408769A CN201410694849.3A CN201410694849A CN104408769A CN 104408769 A CN104408769 A CN 104408769A CN 201410694849 A CN201410694849 A CN 201410694849A CN 104408769 A CN104408769 A CN 104408769A
- Authority
- CN
- China
- Prior art keywords
- dimensional face
- virtual
- face
- dimensional
- cloud server
- 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
Landscapes
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a virtual netmeeting method based on three-dimensional face recognition. The virtual netmeeting method includes S01, establishing a virtual 3D meeting hall; S02, establishing a cloud server; S03, subjecting a 3D face image collecting terminal to collecting real face images of participants in the real time manner; S04, subjecting the cloud server to receiving the 3D face images; S05, performing 3D face image preprocessing, to be specific, subjecting the 3D face images to noise preprocessing, and converting the 3D face images into graying images; S06, extracting the 3D faces, and extracting heads and shoulders from original 3D surfaces; S07, positioning face mark points; S08, registering the 3D faces; S09, subjecting the cloud server to returning face color graphs restored to a virtual meeting display terminal. The virtual netmeeting method has the advantages that 3D face feature recognition is facilitated by the cloud server terminal; participant images are displayed in the virtual netmeeting hall real time, transmission calculation efficiency is high, responding is fast, and application and promotion prospect is bright.
Description
Technical field
The invention belongs to airport intelligence field of safety check, particularly relate to a kind of virtual network conference method based on three-dimensional face identification.
Background technology
At present, virtual network meeting is generally merely able to realize audio frequency and propagates, when using video transmission, because transmission of video images data are comparatively large, video image display effect is poor, and frame losing is serious, prior art, virtual network meeting generally only can effigy, uses specific image or head portrait display participant, can not show the real life facial expression of participant in real time.
Summary of the invention
In order to solve prior art problem, the invention provides and particularly relate to a kind of virtual network conference method based on three-dimensional face identification, three-dimensional face features's identification is carried out by Cloud Server terminal, participant's image is shown in real time in virtual network convention hall, transmission counting yield is high, soon corresponding, can apply.
The technical solution used in the present invention is: based on the virtual network conference method of three-dimensional face identification, comprise the following steps,
S 01, sets up virtual 3D convention hall, for convention hall that is virtually reality like reality, comprises virtual seat, virtual conference table;
S02, sets up Cloud Server, and Cloud Server connects real-time three-dimensional man face image acquiring terminal and virtual meeting display terminal;
S03, real-time three-dimensional man face image acquiring terminal gathers personnel participating in the meeting's real human face image;
S04, three-dimensional face images described in cloud server;
Step S04, after the three-dimensional face images that cloud server image acquisition terminal sends, three-dimensional face images is converted to the computing unit that a series of data are passed to Cloud Server, described computing unit carries out three-dimensional face images pre-service, the extraction of three-dimensional face and three-dimensional face registration.The present invention is based on Cloud Server transmission three-dimensional face images and carry out three-dimensional face identification in cloud server end, three-dimensional face images converts the computing unit that a series of data are passed to Cloud Server to, be cloud data in three-dimensional face images transmitting procedure, transfer efficiency is high, and frame losing phenomenon is few.
S05, three-dimensional face images pre-service: noise preprocess method is carried out to three-dimensional face images, and is converted into gray level image;
The pre-service of step S05 three-dimensional face images specifically comprises the following steps:
(501) denoising is carried out to three-dimensional face images, adopt the triangular surface interpolation method based on cloud data, surface fitting and optimal reconfiguration are carried out to three-dimensional face images cloud data;
(502) use the level and smooth and Taubin method of Laplce to the smoothing process of model;
(503) mesh segmentation is carried out and attitude normalized to facial image, obtain three-dimensional face model.
Image semantic classification filter out noise, display frame is clear, and is conducive to three-dimensional face extraction.
S06, the extraction of three-dimensional face, isolates head and shoulder from initial three-dimensional curved surface;
Step S06 specifically comprises the following steps:
(601) from initial three-dimensional curved surface, isolate head and shoulder, use PCA method to obtain the estimation of attitude, remove shoulder regions;
(602) grid clipping: face is transformed into frontal state by the pre-method for registering based on monumented point, then uses a clipping rectangle face in XOY plane.
The virtual 3D convention hall of the present invention only shows three-dimensional face images in real time, and for facial expression display, therefore, image only retains face parts of images.
S07, face's sign location;
Step S07 specifically comprises the following steps:
(701) two inner eye corners, two tail of the eyes and prenasales are located;
(702) carry out virtual borderlines, obtain the coordinate position of five monumented points (two inner eye corners, two tail of the eyes and prenasales);
(703) based on the point positioning method of continuous Shape Index and geometrical constraint, based on the three-dimensional data of triangle gridding form, by neighborhood of a point matching quadric surface, calculate Shape Index, obtain robust features;
(704) by geometrical constraint, the sign location that attitude is irrelevant is realized.
S08, three-dimensional face registration: according to the monumented point of step S07 location, after registration three-dimensional face, revert to face coloured picture;
Step S08 specifically comprises the following steps: the monumented point behind step of registration S07 location, realizes the rough registration based on monumented point; When three-dimensional face images mark region is lost, monumented point cannot be stablized extraction and carry out three-dimensional face registration by the rough registration method based on Principal Axes Analysis.
After Facial Image Alignment, revert to coloured picture display, meet virtual network meeting display demand.
S09, the face coloured picture result that step S08 recovers by Cloud Server returns virtual meeting display terminal.
Compared with prior art, beneficial effect of the present invention comprises:
The invention provides a kind of virtual network conference method based on three-dimensional face identification, carry out three-dimensional face features's identification by Cloud Server terminal, show participant's image in real time in virtual network convention hall, transmission counting yield is high, soon corresponding, can apply.
Embodiment
Below in conjunction with specific embodiment, the present invention is further described.
Based on the virtual network conference method of three-dimensional face identification, comprise the following steps,
S 01, sets up virtual 3D convention hall, for convention hall that is virtually reality like reality, comprises virtual seat, virtual conference table;
S02, sets up Cloud Server, and Cloud Server connects real-time three-dimensional man face image acquiring terminal and virtual meeting display terminal;
S03, real-time three-dimensional man face image acquiring terminal gathers personnel participating in the meeting's real human face image;
S04, three-dimensional face images described in cloud server;
Step S04, after the three-dimensional face images that cloud server image acquisition terminal sends, three-dimensional face images is converted to the computing unit that a series of data are passed to Cloud Server, described computing unit carries out three-dimensional face images pre-service, the extraction of three-dimensional face and three-dimensional face registration.
S05, three-dimensional face images pre-service: noise preprocess method is carried out to three-dimensional face images, and is converted into gray level image;
The pre-service of step S05 three-dimensional face images specifically comprises the following steps:
(501) denoising is carried out to three-dimensional face images, adopt the triangular surface interpolation method based on cloud data, surface fitting and optimal reconfiguration are carried out to three-dimensional face images cloud data;
(502) use the level and smooth and Taubin method of Laplce to the smoothing process of model;
(503) mesh segmentation is carried out and attitude normalized to facial image, obtain three-dimensional face model.
S06, the extraction of three-dimensional face, isolates head and shoulder from initial three-dimensional curved surface;
Step S06 specifically comprises the following steps:
(601) from initial three-dimensional curved surface, isolate head and shoulder, use PCA method to obtain the estimation of attitude, remove shoulder regions;
(602) grid clipping: face is transformed into frontal state by the pre-method for registering based on monumented point, then uses a clipping rectangle face in XOY plane.
S07, face's sign location;
Step S07 specifically comprises the following steps:
(701) two inner eye corners, two tail of the eyes and prenasales are located;
(702) carry out virtual borderlines, obtain the coordinate position of five monumented points, two inner eye corners, two tail of the eyes and prenasale;
(703) based on the point positioning method of continuous Shape Index and geometrical constraint, based on the three-dimensional data of triangle gridding form, by neighborhood of a point matching quadric surface, calculate Shape Index, obtain robust features;
(704) by geometrical constraint, the sign location that attitude is irrelevant is realized.
S08, three-dimensional face registration: according to the monumented point of step S07 location, after registration three-dimensional face, revert to face coloured picture;
Step S08 specifically comprises the following steps: the monumented point behind step of registration S07 location, realizes the rough registration based on monumented point; When three-dimensional face images mark region is lost, monumented point cannot be stablized extraction and carry out three-dimensional face registration by the rough registration method based on Principal Axes Analysis.
S09, the face coloured picture result that step S08 recovers by Cloud Server returns virtual meeting display terminal.
Below be only the preferred embodiment of the present invention; be noted that for those skilled in the art; under the premise without departing from the principles of the invention, can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (6)
1., based on the virtual network conference method of three-dimensional face identification, it is characterized in that, comprise the following steps,
S 01, sets up virtual 3D convention hall, for convention hall that is virtually reality like reality, comprises virtual seat, virtual conference table;
S02, sets up Cloud Server, and Cloud Server connects real-time three-dimensional man face image acquiring terminal and virtual meeting display terminal;
S03, real-time three-dimensional man face image acquiring terminal gathers personnel participating in the meeting's real human face image;
S04, three-dimensional face images described in cloud server;
S05, three-dimensional face images pre-service: noise preprocess method is carried out to three-dimensional face images, and is converted into gray level image;
S06, the extraction of three-dimensional face, isolates head and shoulder from initial three-dimensional curved surface;
S07, face's sign location;
S08, three-dimensional face registration: according to the monumented point of step S07 location, after registration three-dimensional face, revert to face coloured picture;
S09, the face coloured picture result that step S08 recovers by Cloud Server returns virtual meeting display terminal.
2. the virtual network conference method based on three-dimensional face identification according to claim 1, is characterized in that,
Step S04, after the three-dimensional face images that cloud server image acquisition terminal sends, three-dimensional face images is converted to the computing unit that a series of data are passed to Cloud Server, described computing unit carries out three-dimensional face images pre-service, the extraction of three-dimensional face and three-dimensional face registration.
3. the virtual network conference method based on three-dimensional face identification according to claim 1, is characterized in that,
The pre-service of step S05 three-dimensional face images specifically comprises the following steps:
(501) denoising is carried out to three-dimensional face images, adopt the triangular surface interpolation method based on cloud data, surface fitting and optimal reconfiguration are carried out to three-dimensional face images cloud data;
(502) use the level and smooth and Taubin method of Laplce to the smoothing process of model;
(503) mesh segmentation is carried out and attitude normalized to facial image, obtain three-dimensional face model.
4. the virtual network conference method based on three-dimensional face identification according to claim 1, is characterized in that,
Step S06 specifically comprises the following steps:
(601) from initial three-dimensional curved surface, isolate head and shoulder, use PCA method to obtain the estimation of attitude, remove shoulder regions;
(602) grid clipping: face is transformed into frontal state by the pre-method for registering based on monumented point, then uses a clipping rectangle face in XOY plane.
5. the virtual network conference method based on three-dimensional face identification according to claim 1, is characterized in that,
Step S07 specifically comprises the following steps:
(701) two inner eye corners, two tail of the eyes and prenasales are located;
(702) carry out virtual borderlines, obtain the coordinate position of five monumented points;
(703) based on the point positioning method of continuous Shape Index and geometrical constraint, based on the three-dimensional data of triangle gridding form, by neighborhood of a point matching quadric surface, calculate Shape Index, obtain robust features;
(704) by geometrical constraint, the sign location that attitude is irrelevant is realized.
6. the virtual network conference method based on three-dimensional face identification according to claim 1, it is characterized in that, S08 specifically comprises the following steps: the monumented point behind step of registration S07 location, realizes the rough registration based on monumented point; When three-dimensional face images mark region is lost, monumented point cannot be stablized extraction and carry out three-dimensional face registration by the rough registration method based on Principal Axes Analysis.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410694849.3A CN104408769A (en) | 2014-11-27 | 2014-11-27 | Virtual netmeeting method based on three-dimensional face recognition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410694849.3A CN104408769A (en) | 2014-11-27 | 2014-11-27 | Virtual netmeeting method based on three-dimensional face recognition |
Publications (1)
Publication Number | Publication Date |
---|---|
CN104408769A true CN104408769A (en) | 2015-03-11 |
Family
ID=52646397
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410694849.3A Pending CN104408769A (en) | 2014-11-27 | 2014-11-27 | Virtual netmeeting method based on three-dimensional face recognition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104408769A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109274924A (en) * | 2018-12-07 | 2019-01-25 | 合肥万户网络技术有限公司 | A kind of video-meeting method based on virtual reality |
CN113014852A (en) * | 2019-12-19 | 2021-06-22 | 斑马智行网络(香港)有限公司 | Information prompting method, device and equipment |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1797474A (en) * | 2004-12-30 | 2006-07-05 | 中国科学院自动化研究所 | Fast method for posting players to electronic game |
CN101324961A (en) * | 2008-07-25 | 2008-12-17 | 上海久游网络科技有限公司 | Human face portion three-dimensional picture pasting method in computer virtual world |
CN101814192A (en) * | 2009-02-20 | 2010-08-25 | 三星电子株式会社 | Method for rebuilding real 3D face |
US20110148868A1 (en) * | 2009-12-21 | 2011-06-23 | Electronics And Telecommunications Research Institute | Apparatus and method for reconstructing three-dimensional face avatar through stereo vision and face detection |
US20130301906A1 (en) * | 2012-05-11 | 2013-11-14 | Electronics And Telecommunications Research Institute | Apparatus and method for reconstructing three dimensional faces based on multiple cameras |
CN104091162A (en) * | 2014-07-17 | 2014-10-08 | 东南大学 | Three-dimensional face recognition method based on feature points |
-
2014
- 2014-11-27 CN CN201410694849.3A patent/CN104408769A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1797474A (en) * | 2004-12-30 | 2006-07-05 | 中国科学院自动化研究所 | Fast method for posting players to electronic game |
CN101324961A (en) * | 2008-07-25 | 2008-12-17 | 上海久游网络科技有限公司 | Human face portion three-dimensional picture pasting method in computer virtual world |
CN101814192A (en) * | 2009-02-20 | 2010-08-25 | 三星电子株式会社 | Method for rebuilding real 3D face |
US20110148868A1 (en) * | 2009-12-21 | 2011-06-23 | Electronics And Telecommunications Research Institute | Apparatus and method for reconstructing three-dimensional face avatar through stereo vision and face detection |
US20130301906A1 (en) * | 2012-05-11 | 2013-11-14 | Electronics And Telecommunications Research Institute | Apparatus and method for reconstructing three dimensional faces based on multiple cameras |
CN104091162A (en) * | 2014-07-17 | 2014-10-08 | 东南大学 | Three-dimensional face recognition method based on feature points |
Non-Patent Citations (1)
Title |
---|
杨宝民: "《分布式虚拟现实技术及其应用》", 30 June 2000 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109274924A (en) * | 2018-12-07 | 2019-01-25 | 合肥万户网络技术有限公司 | A kind of video-meeting method based on virtual reality |
CN113014852A (en) * | 2019-12-19 | 2021-06-22 | 斑马智行网络(香港)有限公司 | Information prompting method, device and equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106327571B (en) | A kind of three-dimensional face modeling method and device | |
CN106815555B (en) | Augmented reality method and system for distributed scene target recognition | |
CN104599284B (en) | Three-dimensional facial reconstruction method based on various visual angles mobile phone auto heterodyne image | |
CN102271241A (en) | Image communication method and system based on facial expression/action recognition | |
WO2020133862A1 (en) | Method and apparatus for generating game character model, processor, and terminal | |
CN105956552B (en) | A kind of face blacklist monitoring method | |
CN105678702B (en) | A kind of the human face image sequence generation method and device of feature based tracking | |
CN103679147A (en) | Method and device for identifying model of mobile phone | |
CN103400142B (en) | A kind of pedestrian counting method | |
CN104484803A (en) | Mobile phone payment method employing three-dimensional human face recognition based on neural network | |
CN103327359B (en) | Video significance region searching method applied to video quality evaluation | |
WO2022237249A1 (en) | Three-dimensional reconstruction method, apparatus and system, medium, and computer device | |
CN104966433A (en) | Intelligent glasses assisting deaf-mute conversation | |
CN104169842A (en) | Method for control of video interface, method for operation of video interface, face orientation detector, and video conferencing server | |
CN103745206A (en) | Human face identification method and system | |
CN103927515A (en) | Three-dimension dynamic facial comparison method | |
CN104484669A (en) | Mobile phone payment method based on three-dimensional human face recognition | |
CN103745209A (en) | Human face identification method and system | |
CN104408769A (en) | Virtual netmeeting method based on three-dimensional face recognition | |
CN104898971A (en) | Mouse pointer control method and system based on gaze tracking technology | |
CN103473807A (en) | 3D model transformation system and method | |
Ming et al. | A unified 3D face authentication framework based on robust local mesh SIFT feature | |
CN104408616A (en) | Supermarket prepayment method based on three-dimensional face recognition | |
CN106600702A (en) | Image processing device based on virtual reality | |
Kuang et al. | An effective skeleton extraction method based on Kinect depth image |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into 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: 20150311 |