Multi-Task Learning with Low Rank Attribute Embedding for Person Re-identification

Chi Su1*, Fan Yang2*, Shiliang Zhang1, Qi Tian3, Larry S. Davis2 and Wen Gao1

1Peking University, China
2University of Maryland College Park, United States
3University of Texas at San Antonio, United States

Abstract

We propose a novel Multi-Task Learning with Low Rank Attribute Embedding (MTL-LORAE) framework for person re-identification. Re-identifications from multiple cameras are regarded as related tasks to exploit shared information to improve re-identification accuracy. Both low level fea- tures and semantic/data-driven attributes are utilized. Since attributes are generally correlated, we introduce a low rank attribute embedding into the MTL formulation to embed original binary attributes to a continuous attribute space, where incorrect and incomplete attributes are rectified and recovered to better describe people. The learning objective function consists of a quadratic loss regarding class labels and an attribute embedding error, which is solved by an al- ternating optimization procedure. Experiments on four per- son re-identification datasets have demonstrated that MTL- LORAE outperforms existing approaches by a large margin and produces promising results.

Experimental Results

To be uploaded.

Code

To be uploaded.

Publications