Automatic Tracking and Labeling of Human Activities in a Video Sequence
作者： Ram Nevatia , Jinman Kang , Isaac Cohen , Gérard Medioni , Fengjun Lv
DOI:
关键词: Statistical learning 、 Rigid transformation 、 Invariant (mathematics) 、 Discriminative model 、 Pattern recognition 、 Constant false alarm rate 、 Video sequence 、 Computer science 、 Dynamic programming 、 Artificial intelligence 、 Active appearance model
摘要: This paper presents a novel approach for tracking multiple objects and statistical learning detection of human activities in video sequence. For the tracking, rigid transformation invariant appearance model combining color edge information detected blob is proposed. activity detection, each label regarded as hypothesis. Given some labeled sequences, group features are first extracted from motion trajectories object likelihood feature under that hypothesis calculated. A dynamic programming-based training algorithm applied to get an optimal classifier feature. Then it selects classifiers with most discriminative power combines them form stronger classifier. complies criterion so guaranteed achieve specified rate well minimized false alarm rate. Results on dataset 1show effectiveness proposed algorithm.
usc.edu参考文章(17)
TH Cormen, RL Rivest, CE Leiserson, C Stein, Introduction to Algorithms, 2nd edition. ,(2001)
Harry L. Van Trees, Detection, Estimation, And Modulation Theory ,(1968)
Harry L. Van Trees, Detection, Estimation and Modulation Theory - Part III ,(1971)
J. Yamato, J. Ohya, K. Ishii, Recognizing human action in time-sequential images using hidden Markov model computer vision and pattern recognition. pp. 379- 385 ,(1992) , 10.1109/CVPR.1992.223161
H. Buxton, Advanced visual surveillance using Bayesian networks IEE Colloquium on Image Processing for Security Applications. pp. 9- 9 ,(1997) , 10.1049/IC:19970385
W.A. Hashlamoun, P.K. Varshney, V.N.S. Samarasooriya, A tight upper bound on the Bayesian probability of error IEEE Transactions on Pattern Analysis and Machine Intelligence. ,vol. 16, pp. 220- 224 ,(1994) , 10.1109/34.273728
Somboon Hongeng, Ramakant Nevatia, Large-scale event detection using semi-hidden Markov models international conference on computer vision. pp. 1455- 1462 ,(2003) , 10.1109/ICCV.2003.1238661
Yifan Shi, Aaron F. Bobick, P-Net: A Representation for Partially-Sequenced, Multi-stream Activity 2003 Conference on Computer Vision and Pattern Recognition Workshop. ,vol. 4, pp. 40- 40 ,(2003) , 10.1109/CVPRW.2003.10037
Jinman Kang, I. Cohen, G. Medioni, Continuous tracking within and across camera streams computer vision and pattern recognition. ,vol. 1, pp. 267- 272 ,(2003) , 10.1109/CVPR.2003.1211363
I. Cohen, H. Li, Inference of human postures by classification of 3D human body shape international soi conference. pp. 74- 81 ,(2003) , 10.1109/AMFG.2003.1240827