Person Re-Identification: What Features are Important?

State-of-the-art person re-identication methods seek robust person matching through combining various feature types. Often, these features are implicitly assigned with a single vector of global weights, which are assumed to be universally good for all individuals, independent to their dierent appearances. In this study, we show that certain features play more important role than others under dierent circumstances. Consequently, we propose a novel unsupervised approach for learning a bottom-up feature importance, so features extracted from different individuals are weighted adaptively driven by their unique and inherent appearance attributes. Extensive experiments on two public datasets demonstrate that attribute-sensitive feature importance facilitates more accurate person matching when it is fused together with global weights obtained using existing methods.

• Type: Conference
• Year: 2012
• Where: ECCV, International Workshop on Re-Identification
• Authors: Chunxiao Liu, Shaogang Gong, Chen Change Loy, Xinggang Lin
• Keywords: Multicamera, Person Re-identification, Random Forest, Security, Surveillance, Video Analysis