Shuai Zheng TNT group meeting 1122011 Paper Tracking

� Paper Tracking � Robust view transformation model for gait recognition

� Context-aware fusion: A case study on fusion of gait and face for human

� Effective Unconstrained Face Recognition by Combining Multiple Descriptors and Learned Background Statistics. 2010,

� Cost-sensitive Face Recognition, Zhi-Hua Zhou, PAMI, 2010. Comments: Good motivation: False identification, false

Shuai Zheng, Junge Zhang, Kaiqi Huang, Tieniu Tan, Ran He.

� Motivation �Motivation from related work � Introduction � Experimental results � Conclusions and

� Robust gait representation should be robust to appearance variation caused by the change

� Shared gait representation subspace should be assumed as low-rank. Related Work Handmade Low-Rank

� We present a Robust View Transformation model and Partial Least Square feature selection

GEI from different views Optimized GEI = Low-rank appx A + Sparse error E

A Bag? Remov e it as noise. A overcoat? Remove it as noise. See?

� The proposed method achieves significant performance on the multi-view gait recognition dataset with

sequel � How about the improved low-rank method for other challenge gait recognition dataset?

Slides: 24

Download presentation

Shuai Zheng TNT group meeting 1/12/2011

� Paper Tracking � Robust view transformation model for gait recognition

� Context-aware fusion: A case study on fusion of gait and face for human identification in video, 2010, Pattern Recognition. Comments: This paper introduce how to combine multi biometrics in context-aware way. Great summary for the existing work. New trends in long distance biometrics.

� Effective Unconstrained Face Recognition by Combining Multiple Descriptors and Learned Background Statistics. 2010, PAMI. Comments: How to write a experimental paper? That’s a model.

� Cost-sensitive Face Recognition, Zhi-Hua Zhou, PAMI, 2010. Comments: Good motivation: False identification, false rejection, false acceptance are three different criteria, how to consider the whole cases together? To reduce the expectation of whole cost? Multiclass cost-sensitive KLR seems the point of the paper.

Shuai Zheng, Junge Zhang, Kaiqi Huang, Tieniu Tan, Ran He.

� Motivation �Motivation from related work � Introduction � Experimental results � Conclusions and Future work

� Robust gait representation should be robust to appearance variation caused by the change in viewing angle, carrying or wearing condition.

� Shared gait representation subspace should be assumed as low-rank. Related Work Handmade Low-Rank Truncated Singular Decomposition (TSVD) seems achieved better than original SVD in recent papers on multi-view gait recognition. Robust low-rank method achieved exciting performance in background modeling, face recognition.

� We present a Robust View Transformation model and Partial Least Square feature selection algorithm for multi-view gait recognition.

GEI from different views Optimized GEI = Low-rank appx A + Sparse error E

GEI

A Bag? Remov e it as noise. A overcoat? Remove it as noise. See? What a impressive results of robust View Transformation model for gait representation!

� The proposed method achieves significant performance on the multi-view gait recognition dataset with additional variations caused by wearing or carrying condition change.

sequel � How about the improved low-rank method for other challenge gait recognition dataset? � How about that for visual surveillance system? � Can we achieve super gait recognition? Achieved 99% recognition rates at any viewing angle? How about combine the method with rectified method?

No question? no reward!~