计算机科学 ›› 2015, Vol. 42 ›› Issue (10): 301-305.
邓健康,杨静,孙玉宝,刘青山
DENG Jian-kang, YANG Jing, SUN Yu-bao and LIU Qing-shan
摘要: 如何在计算和存储能力受限的移动平台上实现高效的人脸配准是移动平台人脸应用需要解决的关键问题。主要研究了移动平台上的快速人脸配准问题,为了降低配准模型的计算与存储要求,提出了稀疏约束的级联回归模型。该模型采用稀疏性约束学习回归矩阵,不但能够筛选鲁棒的特征,而且模型的存储空间被压缩到原来的5%左右。基于稀疏级联回归模型,进一步构建了移动平台上人脸配准的快速算法。首先,在人脸检测的基础上,利用二值特征快速定位眼角、嘴角和鼻尖的关键点,估计出人脸的姿态,旋正人脸图像;然后,根据人脸的姿态,选择相应的正脸或侧脸模型,进行稀疏约束的级联回归配准,定位人脸关键点。大量实验结果表明,提出的配准方法精度高、速度快、模型小。在三星Note3智能手机上,每幅人脸图像的配准时间在10ms左右,整个apk文件大小仅为4MB,非常适合移动平台的人脸应用。
[1] Hua G,Fu Y,Turk M,et al.Introduction to the special issue on mobile vision[J].International Journal of Computer Vision,2012,6(3):277-279 [2] Shan Shi-guang,Chang Yi-zheng,Gao Wen,et al.Curse of misa-lignment in face recognition:problem and a novel mis-alignment learning solution[C]∥Proceedings of the Automatic Face and Gesture Recognition,2004.Seoul,Korea,2004:314-320 [3] Cootes T F,Taylor C J.Active shape models—‘smart snakes’[C]∥Proceedings of the British Machine Vision Conference,1992.London,England,1992:266-275 [4] Cootes T F,Taylor C J,Cooper D H,et al.Active shape models-their training and application[J].Computer vision and image understanding,1995,1(1):38-59 [5] Cootes T F,Edwards G J,et al.Active appearance models[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2001,3(6):681-685 [6] Matthews I,Baker S.Active appearance models revisited[J].International Journal of Computer Vision,2004,0(2):135-164 [7] Cristinacce D,Cootes T.Automatic feature localisation withconstrained local models[J].Pattern Recognition,2008,1(10):3054-3067 [8] Zhu Xiang-xin,Ramanan D.Face detection,pose estimation,and landmark localization in the wild[C]∥Proceedings of the Computer Vision and Pattern Recognition,2012.Rhode,America,2012:2879-2886 [9] Zhou Feng,Brandt J,Zhe Lin.Exemplar-based graph matching for robust facial landmark localization[C]∥Proceedings of the International Conference on Computer Vision,2013.Sydney,Australia,2013:1025-1032 [10] Yi Sun,Wang Xiao-gang,Tang Xiao-ou.Deep convolutional network cascade for facial point detection[C]∥Proceedings of the Computer Vision and Pattern Recognition,2013.Portland,America,2013:3476-3483 [11] Zhou Er-jin,Fan Hao-qiang,Cao Zhi-min,et al.Extensive Facial Landmark Localization with Coarse-to-Fine Convolutional Network Cascade[C]∥Workshops of the International Conference on Computer Vision,2013.Sydney,Australia,2013:386-391 [12] Piotr D,Welinder P,Perona P.Cascaded pose regression[C]∥Proceedings of the Computer Vision and Pattern Recognition,2010.San Francisco,America,2010:1078-1085 [13] Cao Xu-dong,Wei Yi-chen,Wen Fang,et al.Face alignment by explicit shape regression[C]∥Proceedings of the Computer Vision and Pattern Recognition,2012.Rhode,America,2012:2887-2894 [14] Xiong Xue-han,De T F.Supervised descent method and its applications to face alignment[C]∥Proceedings of the Computer Vision and Pattern Recognition,2013.Portland,America,2013:532-539 [15] Yan Jun-jie,Zhen Lei,Dong Yi,et al.Learn to Combine Multiple Hypotheses for Accurate Face Alignment[C]∥Workshops of the International Conference on Computer Vision,2013.Sydney,Australia,2013:392-396 [16] Burgos A,Xavier P,Perona P,et al.Robust face landmark estimation under occlusion[C]∥Proceedings of the International Conference on Computer Vision.Sydney,Australia,2013:1513-1530 [17] Ren Shao-qing,Cao Xu-dong,Wei Yi-chen,et al.Face Alignment at 3000 FPS via Regressing Local Binary Features[C]∥Proceedings of the Computer Vision and Pattern Recognition.Columbus,America,2014:1685-1692 [18] Hare S,Saffari A,Torr P H S.Efficient online structured output learning for keypoint-based object tracking[C]∥Proceedings of the Computer Vision and Pattern Recognition.Rhode,America,2012:2887-2894 [19] Tibshirani R.Regression shrinkage and selection via the lasso[J].Journal of the Royal Statistical Society,1994,8(1):267-288 [20] Heikkila M,Pietikainen M,Schmid.Description of interest re-gions with local binary patterns[J].Pattern Recognition,2009,2(3):425-436 [21] Gross R,Matthews I,Cohn J,et al.Multi-pie[J].Image and Vision Computing,2010,8(5):807-813 [22] Sagonas C, Tzimiropoulos G,Zafeiriou S,et al.300 Faces in-the-Wild Challenge:The first facial landmark localization Challenge[C]∥Workshops of the International Conference on Computer Vision.Sydney,Australia,2013:397-403 |
No related articles found! |
|