Segmental training scheme for embedded hidden markov model
-
摘要: 针对嵌入式隐Markov模型再学习问题,提出了分段训练方法用于人脸识别:把当前的训练样本看作整体训练样本的一部分,训练结束后存储训练后的模型参数和中间变量;增加新样本后,以当前模型参数作为初始模型参数,用新增样本训练模型,得到新的中间变量,最后将已存储的中间变量和用新样本计算得到的中间变量合成,得到最终的模型.人脸识别实验结果表明了该方法的可行性.Abstract: A segmental scheme to retrain E-HMM(embedded hidden Markov models) for face recognition was presented. The current samples were assumed to be a subset of the whole training samples, after the training process, the E-HMM parameters and the necessary temporary parameters in the parameter re-estimating process were saved for the use of next step. When new training samples were added, the trained E-HMM parameters were chosen as the initial parameters, the E-HMM was retrained based on the new samples and the new temporary parameters were obtained. These temporary parameters were combined with the saved temporary parameters to form the final E-HMM parameters so that one person face was presented. Experiments on face database showed that the segmental training method was effective.
-
Key words:
- hidden markov model /
- segmental training /
- face recognition /
- stochastic modeling
-
[1] Rabiner L. A tutorial on HMM and selected applications in speech recognition[J]. Proc of IEEE, 1989,77(2):257-286 [2] Kuo S, Agazzi O. Keyword spotting in poorly printed documents using pseudo 2-D hidden Markov models[J]. IEEE Trans on PAMI, 1994, 16(8):842-848 [3] Nefian A V, Hayes III H. Maximum likelihood training of the embedded HMM for face detection and recognition Proc of the International Conference on Image Processing. Vancouver, BC, Canada:,2000:33-36 [4] Wallhoff F, Eickeler S, Gigoll G. A comparison of discrete and continuous output modeling techniques for a pseudo-2D hidden Markov model face recognition system Proc of International Conference on Image Processing. Thessaloniki. Greece:,2001:685-688 [5] Liu Xiaoming, Chen Tsuhan. Video-based face recognition using adaptive hidden Markov models Proc of IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Madison, Wisconsin, USA:,2003:340-345 [6] Zhao W. Face recognition:a literature survey . CS-TR-4167R, 2002 [7] AT&T Laboratories Cambridge. ORL Face database . http://www.uk.research.att.com/facedatabase.html,1994-04-01 [8] Turk M, Pentland A. Eigenceface for recognition[J]. Journal of Cognitive Neuoscience,1991,3(3):71-86 [9] Ben-Arie J, Nandy D. A volumetric/iconic frequency domain representation for object with application for pose invariant face recognition[J]. IEEE Trans on PAMI, 1998,20(3):449-457 [10] Lawrence L, Giles S C, Tsoi A C, et al. Face recognition:a convolutional neural network approach[J]. IEEE Trans on NN, 1997, 8(1):98-113 -
点击查看大图
计量
- 文章访问数: 3246
- HTML全文浏览量: 230
- PDF下载量: 1025
- 被引次数: 0
下载:
