计算机科学 ›› 2017, Vol. 44 ›› Issue (Z11): 263-266.doi: 10.11896/j.issn.1002-137X.2017.11A.055
程载和
CHENG Zai-he
摘要: 为了减轻人脸识别中表情以及姿态等因素变化对识别结果的影响,Xu提出了利用原始样本和对称样本的两步人脸识别算法。但当人脸图像受外在因素干扰产生较大变化时,该方法的识别结果并不理想。因此提出了一种基于因素分解模型的两步人脸识别算法。新算法在特征提取过程中利用因素分解模型将“身份因素”和“表情因素”从人脸图像中分离出来,加以控制。然后提取测试集图像中的新身份和新表情,并将其与训练集中的旧身份或旧表情相互作用,合成新的人脸图像。同时为了保证分类精度,在识别阶段针对原始样本和合成样本分别采用两步人脸识别的方法,充分利用了分数层次融合的优势,进一步提高了算法的识别效果。
[1] SEITZ S M,DYER C R.View morphing[C]∥Proceedings of Computer Graphics.1996:21-30. [2] LANITIS A,TAYLOR C J,COOTES T F.Automatic interpretation and coding of face images using flexible models[J].IEEE Transaction on Pattern Analysis and Machine Intelligence,1997,19(7):743-756. [3] COOPER D H,COOTES T F,TAYLOR C J,et al.Active shape models-their training and application[J].Computer Vision and Image Understanding,1995,1(1):38-59. [4] TENENBAUM J B,FREEMAN W T.Separating style and content with bilinear models[J].Neural Computation,2000,12(6):1247-1283. [5] XU Y,ZHU X J,LI Z M,et al.Using the original and ‘symmetrical face’ training samples to perform representation based two-step face recognition[J].Pattern Recognition,2013,6(4):1151-1158. [6] 徐欢.基于双线性模型的人脸表情识别技术[J].计算机与现代化,2014,7(7):89-93. [7] XU Y,ZHANG Z,LU G,et al.Approximately symmetrical face images for image preprocessing in face recognition and sparse representation based classification[J].Pattern Recognition,2016,54(C):68-82. |
No related articles found! |
|