基于多区域融合的表情鲁棒三维人脸识别算法
|
摘要
为了实现三维人脸识别算法对表情变化的鲁棒性,提出一种基于语义对齐的多区域模板融合三维人脸识别算法。首先,为了实现三维人脸在语义上的对齐,将所有三维人脸模型与预定义标准参考模型做稠密对齐。然后,根据人脸表情具有区域性的特点,为了不受限于区域划分的精准度,提出基于多区域模板的相似度预测方法。最后,采用多数投票法将多个分类器的预测结果融合得到最终识别结果。实验结果表明,在FRGC v2.0表情三维人脸数据库上所提算法可以达到98.69%的rank-1识别率,在含有遮挡变化的Bosphorus数据库上该算法达到84.36%的rank-1识别率。
In order to realize the robustness of three-Dimensional(3 D) face recognition algorithm to expression variations, a multi-region template fusion 3 D face recognition algorithm based on semantic alignment was proposed. Firstly, in order to guarantee the semantic alignment of 3 D faces, all the 3 D face models were densely aligned with a pre-defined standard reference 3 D face model. Then, considering the expressions were regional, to be robust to region division, a multi-region template based similarity prediction method was proposed. Finally, all the prediction results of multiple classifiers were fused by majority voting method. The experimental results show that, the proposed algorithm can achieve the rank-1 face recognition rate of 98.69% on FRGC(the Face Recognition Grand Challenge) v2.0 expression 3 D face database and rank-1 face recognition rate of 84.36% on Bosphorus database with occlusion change.
In order to realize the robustness of three-Dimensional(3 D) face recognition algorithm to expression variations, a multi-region template fusion 3 D face recognition algorithm based on semantic alignment was proposed. Firstly, in order to guarantee the semantic alignment of 3 D faces, all the 3 D face models were densely aligned with a pre-defined standard reference 3 D face model. Then, considering the expressions were regional, to be robust to region division, a multi-region template based similarity prediction method was proposed. Finally, all the prediction results of multiple classifiers were fused by majority voting method. The experimental results show that, the proposed algorithm can achieve the rank-1 face recognition rate of 98.69% on FRGC(the Face Recognition Grand Challenge) v2.0 expression 3 D face database and rank-1 face recognition rate of 84.36% on Bosphorus database with occlusion change.
引文
[1]王跃明,潘纲,吴朝晖.三维人脸识别研究综述[J].计算机辅助设计与图形学学报,2008,20(7):819-829.(WANG Y M,PANG,WU Z H.A survey of 3D face recognition[J].Journal of Computer Aided design&Computer Graphics,2008,20(7):819-829.)
[2]Al-OSAIMI F,BENNAMOUN M,MIAN A.Expression-invariant non-rigid 3D face recognition:a robust approach to expression-aware morphing[C]//Proceedings of the 4th International Symposium on3D Data Processing,Visualization,and Transmission.Piscataway,NJ:IEEE,2008:19-26.
[3]MPIPERIS I,MALASSIOTIS S,STRINTZIS M G.Expression-compensated 3D face recognition with geodesically aligned bilinear models[C]//Proceedings of the 2008 IEEE Second International Conference on Biometrics:Theory,Applications and Systems.Piscataway,NJ:IEEE,2008:1-6.
[4]KAUSHIK V D,BUDHWAR A,DUBEY A,et al.An efficient 3Dface recognition algorithm[C]//NTMS 2009:Proceedings of the3rd International Conference on New Technologies,Mobility and Security.Piscataway,NJ:IEEE,2009:259-263.
[5]BRONSTEIN A M,BRONSTEIN M M,KIMMEL R.Expressioninvariant representations of faces[J].IEEE Transactions on Image Processing,2007,16(1):188-197.
[6]JAHANBIN S,CHOI H,LIU Y,et al.Three dimensional face recognition using iso-geodesic and iso-depth curves[C]//BTAS2008:Proceedings of the 2008 IEEE Second International Conference on Biometrics:Theory,Applications and Systems.Piscataway,NJ:IEEE,2008:1-6.
[7]BERRETTI S,DEL BIMBO A,PALA P.3D face recognition using isogeodesic stripes[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(12):2162-2177.
[8]DRIRA H,BEN AMOR B,DAOUDI M,et al.Pose and expression-invariant 3d face recognition using elastic radial curves[C]//Proceedings of the 2010 British Machine Vision Conference.Durham,UK:BMVA Press,2010:1-11.
[9]BERRETTI S,del BIMBO A,PALA P,et al.A set of selected SIFT features for 3D facial expression recognition[C]//ICPR2010:Proceedings of the 2010 20th International Conference on Pattern Recognition.Washington,DC:IEEE Computer Society,2010:4125-4128.
[10]ALYUZ N,GOKBERK B,AKARUN L.Regional registration for expression resistant 3-D face recognition[J].IEEE Transactions on Information Forensics and Security,2010,5(3):425-440.
[11]SMEETS D,KEUSTERMANS J,VANDERMEULEN D,et al.mesh SIFT:Local surface features for 3D face recognition under expression variations and partial data[J].Computer Vision and Image Understanding,2013,117(2):158-169.
[12]GUO Y L,LEI Y J,LIU L,et al.EI3D:Expression-invariant 3Dface recognition based on feature and shape matching[J].Pattern Recognition Letters,2016,83(Part 3):403-412.
[13]EMAMBAKHSH M,EVANS A.Nasal patches and curves for expression-robust 3D face recognition[J].IEEE transactions on Pattern Analysis and Machine Intelligence,2017,39(5):995-1007.
[14]CHANG K I,BOWYER K W,FLYNN P J.Multiple nose region matching for 3D face recognition under varying facial expression[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2006,28(10):1695-1700.
[15]SANG G L,LI J,ZHAO Q J.Pose-invariant face recognition via RGB-D images[J].Computational Intelligence and Neuroscience,2016(3):1-9.
[16]CHEW W J,SENG K P,ANG L M.Nose tip detection on a three-dimensional face range image invariant to head pose[C]//Proceedings of the 2009 International Multi Conference of Engineers and Computer Scientists.Hong Kong:International Association of Engineers,2009:858-862.
[17]SPREEUWERS L J,VELDHUIS R N J,SULTANALI S,et al.Fixed FAR vote fusion of regional facial classifiers[C]//BIOSIG2014:Proceedings of the 2014 International Conference of the BI-Ometrics Special Interest Group.Piscataway,NJ:IEEE,2014:187-194.
[18]尹宝才,孙艳丰,王成章,等.BJUT-3D三维人脸数据库及其处理技术[J].计算机研究与发展,2009,46(6):1009-1018.(YIN B C,SUN Y F,WANG C Z,et al.BJUT-3D large scale3D face database and information processing[J].Journal of Computer Research and Development,2009,46(6):1009-1018.)
[19]AMBERG B,ROMDHANI S,VETTER T.Optimal step nonrigid ICP algorithms for surface registration[C]//CVPR 2007:Proceedings of the 2007 IEEE Conference on Computer Vision and Pattern Recognition.Piscataway,NJ:IEEE,2007:1-8.
[20]WANG Y M,PAN G,WU Z H.3D face recognition in the presence of expression:a guidance-based constraint deformation approach[C]//CVPR 2007:Proceedings of the 2007 Computer Vision and Pattern Recognition.Piscataway,NJ:IEEE,2007:1180-1187.
[21]SAVRAN A,ALYUZ N,DIBEKLIOGLU H,et al.Bosphorus database for 3D face analysis[C]//Proceedings of the 2008 European Workshop on Biometrics and Identity Management,LNCS5372.Berlin:Springer,2008:47-56.
[22]DENG X,DA F P,SHAO H J.Adaptive feature selection based on reconstruction residual and accurately located landmarks for expression-robust 3D face recognition[J].Signal,Image and Video Processing,2017,11(7):1305-1312.
[23]BAGCHI P,BHATTACHARJEE D,NASIPURI M.Robust 3Dface recognition in presence of pose and partial occlusions or missing parts[J].International Journal in Foundations of Computer Science&Technology,2014,4(4):21-35.
[24]GANGULY S,BHATTACHARJEE D,NASIPURI M.Depth based occlusion detection and localization from 3D face image[J].International Journal of Image,Graphics&Signal Processing,2015,7(5):20-31.
[2]Al-OSAIMI F,BENNAMOUN M,MIAN A.Expression-invariant non-rigid 3D face recognition:a robust approach to expression-aware morphing[C]//Proceedings of the 4th International Symposium on3D Data Processing,Visualization,and Transmission.Piscataway,NJ:IEEE,2008:19-26.
[3]MPIPERIS I,MALASSIOTIS S,STRINTZIS M G.Expression-compensated 3D face recognition with geodesically aligned bilinear models[C]//Proceedings of the 2008 IEEE Second International Conference on Biometrics:Theory,Applications and Systems.Piscataway,NJ:IEEE,2008:1-6.
[4]KAUSHIK V D,BUDHWAR A,DUBEY A,et al.An efficient 3Dface recognition algorithm[C]//NTMS 2009:Proceedings of the3rd International Conference on New Technologies,Mobility and Security.Piscataway,NJ:IEEE,2009:259-263.
[5]BRONSTEIN A M,BRONSTEIN M M,KIMMEL R.Expressioninvariant representations of faces[J].IEEE Transactions on Image Processing,2007,16(1):188-197.
[6]JAHANBIN S,CHOI H,LIU Y,et al.Three dimensional face recognition using iso-geodesic and iso-depth curves[C]//BTAS2008:Proceedings of the 2008 IEEE Second International Conference on Biometrics:Theory,Applications and Systems.Piscataway,NJ:IEEE,2008:1-6.
[7]BERRETTI S,DEL BIMBO A,PALA P.3D face recognition using isogeodesic stripes[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2010,32(12):2162-2177.
[8]DRIRA H,BEN AMOR B,DAOUDI M,et al.Pose and expression-invariant 3d face recognition using elastic radial curves[C]//Proceedings of the 2010 British Machine Vision Conference.Durham,UK:BMVA Press,2010:1-11.
[9]BERRETTI S,del BIMBO A,PALA P,et al.A set of selected SIFT features for 3D facial expression recognition[C]//ICPR2010:Proceedings of the 2010 20th International Conference on Pattern Recognition.Washington,DC:IEEE Computer Society,2010:4125-4128.
[10]ALYUZ N,GOKBERK B,AKARUN L.Regional registration for expression resistant 3-D face recognition[J].IEEE Transactions on Information Forensics and Security,2010,5(3):425-440.
[11]SMEETS D,KEUSTERMANS J,VANDERMEULEN D,et al.mesh SIFT:Local surface features for 3D face recognition under expression variations and partial data[J].Computer Vision and Image Understanding,2013,117(2):158-169.
[12]GUO Y L,LEI Y J,LIU L,et al.EI3D:Expression-invariant 3Dface recognition based on feature and shape matching[J].Pattern Recognition Letters,2016,83(Part 3):403-412.
[13]EMAMBAKHSH M,EVANS A.Nasal patches and curves for expression-robust 3D face recognition[J].IEEE transactions on Pattern Analysis and Machine Intelligence,2017,39(5):995-1007.
[14]CHANG K I,BOWYER K W,FLYNN P J.Multiple nose region matching for 3D face recognition under varying facial expression[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2006,28(10):1695-1700.
[15]SANG G L,LI J,ZHAO Q J.Pose-invariant face recognition via RGB-D images[J].Computational Intelligence and Neuroscience,2016(3):1-9.
[16]CHEW W J,SENG K P,ANG L M.Nose tip detection on a three-dimensional face range image invariant to head pose[C]//Proceedings of the 2009 International Multi Conference of Engineers and Computer Scientists.Hong Kong:International Association of Engineers,2009:858-862.
[17]SPREEUWERS L J,VELDHUIS R N J,SULTANALI S,et al.Fixed FAR vote fusion of regional facial classifiers[C]//BIOSIG2014:Proceedings of the 2014 International Conference of the BI-Ometrics Special Interest Group.Piscataway,NJ:IEEE,2014:187-194.
[18]尹宝才,孙艳丰,王成章,等.BJUT-3D三维人脸数据库及其处理技术[J].计算机研究与发展,2009,46(6):1009-1018.(YIN B C,SUN Y F,WANG C Z,et al.BJUT-3D large scale3D face database and information processing[J].Journal of Computer Research and Development,2009,46(6):1009-1018.)
[19]AMBERG B,ROMDHANI S,VETTER T.Optimal step nonrigid ICP algorithms for surface registration[C]//CVPR 2007:Proceedings of the 2007 IEEE Conference on Computer Vision and Pattern Recognition.Piscataway,NJ:IEEE,2007:1-8.
[20]WANG Y M,PAN G,WU Z H.3D face recognition in the presence of expression:a guidance-based constraint deformation approach[C]//CVPR 2007:Proceedings of the 2007 Computer Vision and Pattern Recognition.Piscataway,NJ:IEEE,2007:1180-1187.
[21]SAVRAN A,ALYUZ N,DIBEKLIOGLU H,et al.Bosphorus database for 3D face analysis[C]//Proceedings of the 2008 European Workshop on Biometrics and Identity Management,LNCS5372.Berlin:Springer,2008:47-56.
[22]DENG X,DA F P,SHAO H J.Adaptive feature selection based on reconstruction residual and accurately located landmarks for expression-robust 3D face recognition[J].Signal,Image and Video Processing,2017,11(7):1305-1312.
[23]BAGCHI P,BHATTACHARJEE D,NASIPURI M.Robust 3Dface recognition in presence of pose and partial occlusions or missing parts[J].International Journal in Foundations of Computer Science&Technology,2014,4(4):21-35.
[24]GANGULY S,BHATTACHARJEE D,NASIPURI M.Depth based occlusion detection and localization from 3D face image[J].International Journal of Image,Graphics&Signal Processing,2015,7(5):20-31.