CN103310415A - Face-based defected image inpainting method and system - Google Patents
Face-based defected image inpainting method and system Download PDFInfo
- Publication number
- CN103310415A CN103310415A CN2013100845592A CN201310084559A CN103310415A CN 103310415 A CN103310415 A CN 103310415A CN 2013100845592 A CN2013100845592 A CN 2013100845592A CN 201310084559 A CN201310084559 A CN 201310084559A CN 103310415 A CN103310415 A CN 103310415A
- Authority
- CN
- China
- Prior art keywords
- face
- image
- damaged
- defect
- people
- 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
Images
Landscapes
- Image Processing (AREA)
- Processing Or Creating Images (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a face-based defected image inpainting method and system. The method comprises the following steps of acquiring pixels of a whole face from a defected face image; retrieving a plurality of face images, similar to the defected face image, from a face image database through sparse representation according to the pixels of the defected face image; extracting the best-matched defected portion from the face images according to a defect area; and intercepting the best-matched defected area from the best-matched face image, and fusing the defected portion by using a Laplace equation so as to obtain a repaired image. According to the method provided by the embodiments of the invention, the complexity of data calculation during image retrieval is reduced by using sparse representation, so that the efficiency is increased; and meanwhile, the image of the defect area is subjected to inpainting by using spectrogram information of the image.
Description
Technical field
The present invention relates to technical field of image processing, particularly the damaged benefit of a kind of image based on people's face is painted method and system.
Background technology
Recognition of face is the most natural, and is harmless, user friendly biometric method.Most of existing system can only success identify face under some constraint condition.And some in particular cases its recognition result can be subjected to image influence, for example, sunglasses, beard or shelter in illumination, countenance, posture, visual angle, the especially occlusion.
To the identification of people's face adopt be whole complete facial image as identifying object, when the human face image information of identifying is imperfect, general recognition methods can't be suitable for.
The benefit method of painting of the breakage image that generally uses has at present, paints algorithm and paints algorithm based on the benefit of Markov Random Fields based on the benefit of Poisson equation.But traditional benefit is painted algorithm and more is applicable to the nature picture, not can solve people's face and mends this problem of painting.
There is following shortcoming in traditional benefit method of painting:
(1) can't handle zone to be matched and target area illumination, colour-difference is apart from situations such as significant changes.
(2) be applied to the people on the face, because the disappearance of primitive man's face pictorial information is painted under the algorithm frame in the tradition benefit, can not well finding zonule to be matched, painting to carry out the target benefit.
Summary of the invention
Purpose of the present invention is intended to solve at least one of above-mentioned technological deficiency.
For achieving the above object, the embodiment of one aspect of the present invention proposes the damaged benefit of a kind of image based on people's face and paints method, may further comprise the steps: S1: the pixel of obtaining whole people's face from damaged facial image; S2: the described pixel according to damaged facial image is retrieved a plurality of facial images close with described damaged facial image by rarefaction representation from the facial image database; S3: from described a plurality of facial images, extract the defect of mating most according to defect area; And S4: the defect area that intercepting is mated most from the described facial image that mates is most also utilized Laplace's equation that defect is merged acquisition and is repaired image.
According to the method for the embodiment of the invention, the data computation complexity when utilizing rarefaction representation to reduce retrieving images, thus improved efficient, utilize the spectrogram information of image that the image of defect area is mended simultaneously and paint.
In one embodiment of the present of invention, described step S4 specifically comprises: S41: utilize contiguous method that the local annexation of described defect of mating is most obtained the divergence matrix; S42: obtain Laplce's matrix according to described divergence matrix; And S43: utilize blending algorithm to obtain defect area to be repaired according to described Laplce's matrix.
In one embodiment of the present of invention, when obtaining defect area according to the described defect that retrieves, observe following constraint condition:
Wherein, G
UIt is defect area, G
RIn be the defect that retrieves.
In one embodiment of the invention, described step S41 further comprises: S411: ask for described each pixel of mating most in the defect area as the node of correspondence; S412: be a plurality of pixel nodes of described each pixel node and its periphery matrix that connects; S413: obtain local connection matrix for described annexation arranges the connection weights; And S414: calculate the divergence matrix according to described local connection matrix.
In one embodiment of the invention, described facial image database comprises that various facial images satisfy the search coupling of facial image.
For achieving the above object, embodiments of the invention propose the damaged benefit of a kind of image based on people's face on the other hand and paint system, comprising: acquisition module, for the pixel of obtaining whole people's face from damaged facial image; Retrieval module is used for retrieving with described damaged facial image close a plurality of facial images by rarefaction representation from the facial image database according to the described pixel of damaged facial image; Extraction module is used for extracting the defect of mating most according to defect area from described a plurality of facial images; And Fusion Module, obtain the reparation image for intercepting the defect area of mating most from the described facial image that mates most and utilizing Laplace's equation that defect is merged.
According to the system of the embodiment of the invention, the data computation complexity when utilizing rarefaction representation to reduce retrieving images, thus improved efficient, utilize the spectrogram information of image that the image of defect area is mended simultaneously and paint.
In one embodiment of the present of invention, described Fusion Module specifically comprises: first obtains submodule, is used for utilizing contiguous method that the local annexation of described defect of mating is most obtained the divergence matrix; Second obtains submodule, is used for obtaining Laplce's matrix according to described divergence matrix; And the 3rd obtain submodule, is used for utilizing blending algorithm to obtain defect area to be repaired according to described Laplce's matrix.
In one embodiment of the present of invention, when obtaining defect area according to the described defect that retrieves, observe following constraint condition:
Wherein, G
UIt is defect area, G
RIn be the defect that retrieves.
In one embodiment of the present of invention, described first obtains submodule further comprises: ask for the unit, be used for asking for described each pixel of defect area of mating most as the node of correspondence; Set up the unit, be used to a plurality of pixel nodes of described each pixel node and its periphery to establish a connection; Acquiring unit is used to described annexation that the connection weights are set and obtains local connection matrix; And computing unit, be used for calculating the divergence matrix according to described local connection matrix.
In one embodiment of the invention, described facial image database comprises that various facial images satisfy the search coupling of facial image.
The aspect that the present invention adds and advantage part in the following description provide, and part will become obviously from the following description, or recognize by practice of the present invention.
Description of drawings
Above-mentioned and/or the additional aspect of the present invention and advantage are from obviously and easily understanding becoming the description of embodiment below in conjunction with accompanying drawing, wherein:
Fig. 1 is for to paint the synoptic diagram of method based on the damaged benefit of image of people's face according to an embodiment of the invention;
Fig. 2 is for to paint the process flow diagram of method based on the damaged benefit of image of people's face according to an embodiment of the invention;
Fig. 3 paints figure as a result for mending according to an embodiment of the invention;
Fig. 4 is for to paint the frame diagram of system based on the damaged benefit of image of people's face according to an embodiment of the invention;
Fig. 5 is the frame diagram of Fusion Module according to an embodiment of the invention;
Fig. 6 is first frame diagram that obtains submodule according to an embodiment of the invention; And
Fig. 7 is the result data figure that identification is according to an embodiment of the invention tested.
Embodiment
Describe embodiments of the invention below in detail, the example of embodiment is shown in the drawings, and wherein identical or similar label is represented identical or similar elements or the element with identical or similar functions from start to finish.Be exemplary below by the embodiment that is described with reference to the drawings, only be used for explaining the present invention, and can not be interpreted as limitation of the present invention.
Fig. 1 paints the synoptic diagram of method for the damaged benefit of the image based on people's face of the embodiment of the invention.Fig. 2 is for to paint the process flow diagram of method based on the damaged benefit of image of people's face according to an embodiment of the invention.As shown in Figure 2, paint method according to the damaged benefit of the image based on people's face of the embodiment of the invention, may further comprise the steps:
Step S101 obtains the pixel of whole people's face from damaged facial image.Obtain damaged facial image, and therefrom obtain the pixel of whole people's face in this image.
In one embodiment of the invention, has certain similarity the present invention in design in such cases for the face between the different people.
Step S102 retrieves a plurality of facial images close with damaged facial image according to the pixel of damaged facial image by rarefaction representation from the facial image database.Wherein, the facial image database comprises that various facial images satisfy the search coupling of facial image.
Particularly, according to the pixel of the damaged facial image that obtains and utilize the similar a plurality of similar images of rarefaction representation cluster and damaged image from the facial image database, wherein, the facial image data of facial image database cover each facial image.
Step S103 extracts the defect of mating most according to defect area from a plurality of facial images.
Particularly, at first from damaged image, obtain the data boundary of defect area, for example, pixel etc.From a plurality of similar a plurality of images of retrieving, extract the boundary information that conforms to the defect area border then, for example, pixel etc.According to the border of defect area and the defect border goodness of fit of retrieving images, from a plurality of similar images, choose the most identical similar image.
Step S104, the defect area that intercepting is mated most from the facial image that mates is most also utilized Laplace's equation that defect is merged acquisition and is repaired image.At first, ask for and mate in the defect area each pixel most as the node of correspondence.Then, be 8 pixel nodes of each pixel node and its periphery matrix W that connects.Afterwards, for annexation arranges the connection weights, the link weights be designated as 1, other be designated as 0, thereby obtain local connection matrix.At last, calculate the divergence matrix D according to local connection matrix,
Obtain Laplce's matrix according to the divergence matrix.Wherein, Laplce's defined matrix is L=D-W.
Utilize blending algorithm to obtain defect area to be repaired according to Laplce's matrix.LG
U=LG
R, wherein, G
UBe area information to be repaired, G
RInformation for the defect area that retrieves.In this formula, according to boundary condition, can solve G by finding the solution the mode of linear equation
UIts benefit is painted the result as shown in Figure 3.
According to the method for the embodiment of the invention, the data computation complexity when utilizing rarefaction representation to reduce retrieving images, thus improved efficient, utilize the spectrogram information of image that the image of defect area is mended simultaneously and paint.
Fig. 4 is for to paint the frame diagram of system based on the damaged benefit of image of people's face according to an embodiment of the invention.As shown in Figure 4, the damaged benefit of the image based on the people's face system of painting according to the embodiment of the invention comprises acquisition module 100, retrieval module 200, extraction module 300 and Fusion Module 400.
Particularly, acquisition module 100 is used for obtaining from damaged facial image the pixel of whole people's face.
In one embodiment of the invention, has certain similarity the present invention in design in such cases for the face between the different people.
Retrieval module 200 is used for retrieving with damaged facial image close a plurality of facial images by rarefaction representation from the facial image database according to the pixel of damaged facial image.According to the pixel of the damaged facial image that obtains and utilize the similar a plurality of similar images of rarefaction representation cluster and damaged image from the facial image database, wherein, the facial image data of facial image database cover each facial image.And the facial image database comprises that various facial images satisfy the search coupling of facial image.
Extraction module 300 is used for extracting the defect of mating most according to defect area from a plurality of facial images.At first from damaged image, obtain the data boundary of defect area, for example, pixel etc.From a plurality of similar a plurality of images of retrieving, extract the boundary information that conforms to the defect area border then, for example, pixel etc.According to the border of defect area and the defect border goodness of fit of retrieving images, from a plurality of similar images, choose the most identical similar image.
Fusion Module 400 is for the defect area of mating most from the facial image intercepting of mating most and utilize Laplace's equation that defect is merged acquisition reparation image.
Fig. 5 is the frame diagram of Fusion Module according to an embodiment of the invention.As shown in Figure 5, Fusion Module 400 comprises that first obtains submodule 410, second and obtain submodule 420 and the 3rd and obtain submodule 430.
First obtains submodule 410 for the local annexation acquisition divergence matrix D of the defect of utilizing contiguous method to mate most,
Fig. 6 is first frame diagram that obtains submodule according to an embodiment of the invention.As shown in Figure 6, first obtain submodule 410 specifically comprise ask for unit 411, set up unit 412, acquiring unit 413 and computing unit 414.
Second obtains submodule 420 is used for obtaining Laplce's matrix according to the divergence matrix.Wherein, Laplce's defined matrix is L=D-W.
The 3rd obtains submodule 430 is used for utilizing blending algorithm to obtain defect area to be repaired according to Laplce's matrix.LG
U=LG
R, wherein, G
UBe area information to be repaired, G
RInformation for the defect area that retrieves.In this formula, according to boundary condition, can solve G by finding the solution the mode of linear equation
U
According to the system of the embodiment of the invention, the data computation complexity when utilizing rarefaction representation to reduce retrieving images, thus improved efficient, utilize the spectrogram information of image that the image of defect area is mended simultaneously and paint.
Also carried out the identification test in the present invention, further embodied superiority of the present invention, comparing result is specific as follows.
In one embodiment of the invention, utilizing the benefit propose to paint algorithm goes the benefit of disappearance people's face of finishing to paint and identify.Face recognition algorithms is being carried out at the facial image database.In training, choose 847 and do not block people's face (121subjects, 7images per subject) and be used to.In test, the remaining damaged people's face that has will be classified into two kinds of set.Be designated as AR1 and AR2 respectively.The set of this two class has comprised respectively wears glasses and wears the face that having of scarf blocked.We use the most classical face recognition algorithms of two classes.Namely based on the recognition of face of PCA with based on the face recognition algorithms of Gabor feature.This two classes algorithm is used in test respectively and paints on the sample at damaged sample and benefit.As shown in Figure 5, as can be seen, the present invention mends people's face of painting can improve discrimination greatly from the recognition result contrast.
The specific operation process that should be appreciated that each module in the system embodiment of the present invention and unit can be identical with the description among the method embodiment, is not described in detail herein.
Although illustrated and described embodiments of the invention above, be understandable that, above-described embodiment is exemplary, can not be interpreted as limitation of the present invention, those of ordinary skill in the art can change above-described embodiment under the situation that does not break away from principle of the present invention and aim within the scope of the invention, modification, replacement and modification.
Claims (10)
1. the damaged benefit of the image based on people's face is painted method, it is characterized in that, may further comprise the steps:
S1: the pixel of from damaged facial image, obtaining whole people's face;
S2: the described pixel according to damaged facial image is retrieved a plurality of facial images close with described damaged facial image by rarefaction representation from the facial image database;
S3: from described a plurality of facial images, extract the defect of mating most according to defect area; And
S4: the defect area that intercepting is mated most from the described facial image that mates is most also utilized Laplace's equation that defect is merged acquisition and is repaired image.
2. the damaged benefit of the image based on people's face as claimed in claim 1 is painted method, it is characterized in that described step S4 specifically comprises:
S41: utilize contiguous method that the local annexation of described defect of mating is most obtained the divergence matrix;
S42: obtain Laplce's matrix according to described divergence matrix; And
S43: utilize blending algorithm to obtain defect area to be repaired according to described Laplce's matrix.
3. the damaged benefit of the image based on people's face as claimed in claim 2 is painted method, it is characterized in that, when obtaining defect area according to the described defect that retrieves, observes following constraint condition:
Wherein, G
UIt is defect area, G
RIn be the defect that retrieves.
4. the damaged benefit of the image based on people's face as claimed in claim 2 is painted method, it is characterized in that described step S41 further comprises:
S411: ask for described each pixel of mating most in the defect area as the node of correspondence;
S412: be a plurality of pixel nodes of described each pixel node and its periphery matrix that connects;
S413: obtain local connection matrix for described annexation arranges the connection weights; And
S414: calculate the divergence matrix according to described local connection matrix.
5. paint method as the damaged benefit of each described image based on people's face of claim 1-4, it is characterized in that, described facial image database comprises that various facial images satisfy the search coupling of facial image.
6. the damaged benefit of the image based on people's face is painted system, it is characterized in that, comprising:
Acquisition module is for the pixel of obtaining whole people's face from damaged facial image;
Retrieval module is used for retrieving with described damaged facial image close a plurality of facial images by rarefaction representation from the facial image database according to the described pixel of damaged facial image;
Extraction module is used for extracting the defect of mating most according to defect area from described a plurality of facial images; And
Fusion Module obtains the reparation image for intercepting the defect area of mating most from the described facial image that mates most and utilizing Laplace's equation that defect is merged.
7. the damaged benefit of the image based on people's face as claimed in claim 6 is painted system, it is characterized in that described Fusion Module specifically comprises:
First obtains submodule, is used for utilizing contiguous method that the local annexation of described defect of mating is most obtained the divergence matrix;
Second obtains submodule, is used for obtaining Laplce's matrix according to described divergence matrix; And
The 3rd obtains submodule, is used for utilizing blending algorithm to obtain defect area to be repaired according to described Laplce's matrix.
8. the damaged benefit of the image based on people's face as claimed in claim 7 is painted system, it is characterized in that, when obtaining defect area according to the described defect that retrieves, observes following constraint condition:
Wherein, G
UIt is defect area, G
RIn be the defect that retrieves.
9. the damaged benefit of the image based on people's face as claimed in claim 7 is painted system, it is characterized in that, described first obtains submodule specifically comprises:
Ask for the unit, be used for asking for described each pixel of defect area of mating most as the node of correspondence;
Set up the unit, be used to a plurality of pixel nodes of described each pixel node and its periphery to establish a connection;
Acquiring unit is used to described annexation that the connection weights are set and obtains local connection matrix; And
Computing unit is used for calculating the divergence matrix according to described local connection matrix.
10. paint system as the damaged benefit of each described image based on people's face of claim 6-9, it is characterized in that, described facial image database comprises that various facial images satisfy the search coupling of facial image.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2013100845592A CN103310415A (en) | 2013-03-15 | 2013-03-15 | Face-based defected image inpainting method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2013100845592A CN103310415A (en) | 2013-03-15 | 2013-03-15 | Face-based defected image inpainting method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN103310415A true CN103310415A (en) | 2013-09-18 |
Family
ID=49135593
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2013100845592A Pending CN103310415A (en) | 2013-03-15 | 2013-03-15 | Face-based defected image inpainting method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103310415A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107169934A (en) * | 2017-05-10 | 2017-09-15 | 河海大学 | A kind of image mending method based on different redundant dictionaries |
CN108765315A (en) * | 2018-05-04 | 2018-11-06 | Oppo广东移动通信有限公司 | Image completion method, apparatus, computer equipment and storage medium |
CN108875638A (en) * | 2018-06-20 | 2018-11-23 | 北京京东金融科技控股有限公司 | Face matching test method and device and system |
CN109360170A (en) * | 2018-10-24 | 2019-02-19 | 北京工商大学 | Face restorative procedure based on advanced features |
CN109598210A (en) * | 2018-11-16 | 2019-04-09 | 三星电子(中国)研发中心 | A kind of image processing method and device |
CN109903298A (en) * | 2019-03-12 | 2019-06-18 | 数坤(北京)网络科技有限公司 | Restorative procedure, system and the computer storage medium of blood vessel segmentation image fracture |
CN110020578A (en) * | 2018-01-10 | 2019-07-16 | 广东欧珀移动通信有限公司 | Image processing method, device, storage medium and electronic equipment |
CN113330480A (en) * | 2019-02-11 | 2021-08-31 | 康蒂-特米克微电子有限公司 | Modular image restoration method |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040091137A1 (en) * | 2002-11-04 | 2004-05-13 | Samsung Electronics Co., Ltd. | System and method for detecting face |
US20060291001A1 (en) * | 2005-06-18 | 2006-12-28 | Samsung Electronics Co., Ltd. | Apparatus and method for detecting occluded face and apparatus and method for discriminating illegal user using the same |
CN102855496A (en) * | 2012-08-24 | 2013-01-02 | 苏州大学 | Method and system for authenticating shielded face |
CN102915436A (en) * | 2012-10-25 | 2013-02-06 | 北京邮电大学 | Sparse representation face recognition method based on intra-class variation dictionary and training image |
-
2013
- 2013-03-15 CN CN2013100845592A patent/CN103310415A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040091137A1 (en) * | 2002-11-04 | 2004-05-13 | Samsung Electronics Co., Ltd. | System and method for detecting face |
US20060291001A1 (en) * | 2005-06-18 | 2006-12-28 | Samsung Electronics Co., Ltd. | Apparatus and method for detecting occluded face and apparatus and method for discriminating illegal user using the same |
CN102855496A (en) * | 2012-08-24 | 2013-01-02 | 苏州大学 | Method and system for authenticating shielded face |
CN102915436A (en) * | 2012-10-25 | 2013-02-06 | 北京邮电大学 | Sparse representation face recognition method based on intra-class variation dictionary and training image |
Non-Patent Citations (1)
Title |
---|
YUE DENG ET AL.: "Graph Laplace for Occluded Face Completion and Recognition", 《IEEE TRANSACTIONS ON IMAGE PROCESSING》, vol. 20, no. 8, 31 August 2011 (2011-08-31), pages 2329 - 2338, XP011411840, DOI: doi:10.1109/TIP.2011.2109729 * |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107169934A (en) * | 2017-05-10 | 2017-09-15 | 河海大学 | A kind of image mending method based on different redundant dictionaries |
US11386699B2 (en) | 2018-01-10 | 2022-07-12 | Guangdong Oppo Mobile Telecommunications Corp., Ltd. | Image processing method, apparatus, storage medium, and electronic device |
CN110020578A (en) * | 2018-01-10 | 2019-07-16 | 广东欧珀移动通信有限公司 | Image processing method, device, storage medium and electronic equipment |
CN108765315A (en) * | 2018-05-04 | 2018-11-06 | Oppo广东移动通信有限公司 | Image completion method, apparatus, computer equipment and storage medium |
CN108765315B (en) * | 2018-05-04 | 2021-09-07 | Oppo广东移动通信有限公司 | Image completion method and device, computer equipment and storage medium |
CN108875638B (en) * | 2018-06-20 | 2020-07-31 | 京东数字科技控股有限公司 | Face matching test method, device and system |
CN108875638A (en) * | 2018-06-20 | 2018-11-23 | 北京京东金融科技控股有限公司 | Face matching test method and device and system |
CN109360170A (en) * | 2018-10-24 | 2019-02-19 | 北京工商大学 | Face restorative procedure based on advanced features |
CN109598210B (en) * | 2018-11-16 | 2020-10-27 | 三星电子(中国)研发中心 | Picture processing method and device |
CN109598210A (en) * | 2018-11-16 | 2019-04-09 | 三星电子(中国)研发中心 | A kind of image processing method and device |
CN113330480A (en) * | 2019-02-11 | 2021-08-31 | 康蒂-特米克微电子有限公司 | Modular image restoration method |
US11961215B2 (en) | 2019-02-11 | 2024-04-16 | Conti Temic Microelectronic Gmbh | Modular inpainting method |
CN109903298B (en) * | 2019-03-12 | 2021-03-02 | 数坤(北京)网络科技有限公司 | Method, system and computer storage medium for repairing blood vessel segmentation image fracture |
CN109903298A (en) * | 2019-03-12 | 2019-06-18 | 数坤(北京)网络科技有限公司 | Restorative procedure, system and the computer storage medium of blood vessel segmentation image fracture |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103310415A (en) | Face-based defected image inpainting method and system | |
US9530071B2 (en) | Hierarchical interlinked multi-scale convolutional network for image parsing | |
Shreve et al. | Macro-and micro-expression spotting in long videos using spatio-temporal strain | |
WO2017016515A1 (en) | Key frame acquisition method for human image video system | |
CN105354558B (en) | Humanface image matching method | |
Lin et al. | Super-resolved faces for improved face recognition from surveillance video | |
JP2007317062A (en) | Person recognition apparatus and method | |
JP6410450B2 (en) | Object identification device, object identification method, and program | |
Ouanan et al. | Facial landmark localization: Past, present and future | |
Chakravarty et al. | Coupled sparse dictionary for depth-based cup segmentation from single color fundus image | |
Gawali et al. | 3d face recognition using geodesic facial curves to handle expression, occlusion and pose variations | |
JP2007300185A (en) | Image monitoring apparatus | |
Torrisi et al. | Selecting discriminative CLBP patterns for age estimation | |
Devadethan et al. | Face detection and facial feature extraction based on a fusion of knowledge based method and morphological image processing | |
JP3577908B2 (en) | Face image recognition system | |
JP2013218605A (en) | Image recognition device, image recognition method, and program | |
Kauser et al. | Automatic facial expression recognition: A survey based on feature extraction and classification techniques | |
Wang et al. | Couple metric learning based on separable criteria with its application in cross-view gait recognition | |
Hu et al. | Patch-based face recognition from video | |
Majeed et al. | Nose tip detection in 3D face image based on maximum intensity algorithm | |
Poursaberi et al. | Modified multiscale vesselness filter for facial feature detection | |
Dixit et al. | A hybrid approach of face recognition using bezier curve | |
Khaparde et al. | Face detection using color based segmentation and morphological processing–a case study | |
Heravi et al. | A Morphable Model to simulate rejuvenation trajectory of 3D face images: Preliminary results | |
Wei et al. | Automatic facial pose determination of 3D range data for face model and expression identification |
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 |
Application publication date: 20130918 |
|
RJ01 | Rejection of invention patent application after publication |