Deep attention network for person re-identification with multi-loss

Abstract

Person re-identification (person re-ID) is one of the most challenging tasks in the computer vision area as it involves large variations in human appearances, human poses, background illuminations, camera views, etc. In particular, images for person re-ID are mostly low resolution due to the long-range deployment of the cameras and the cropping operation from the surveillance system. In this paper, we present a novel deep Siamese person re-ID network equipped with an attention mechanism, constrained by a multi-loss function. The attention mechanism enhances the discriminability of the network by emphasizing effective features and suppressing the less useful ones. The purpose of the multi-loss function is to diminish distances of identical persons and at the same time expand distances between dissimilar persons in the learned feature space. Extensive comparative evaluations demonstrate that the proposed method significantly outperforms a number of state-of-the-art approaches, including both conventional and deep network based ones, on the challenging Market1501 and CUHK03 data sets.