Person re-identification by localizing discriminative regions

Abstract

Person re-identification is a challenging task of matching a person’s image across multiple images captured from different camera views. Recently, deep learning based approaches have been proposed that show promising performance on this task. However, most of these approaches use whole image features to compute the similarity between images. This is not very intuitive since not all the regions in an image contain information about the person identity. In this paper, we introduce an end-to-end Siamese convolutional neural network that firstly localizes discriminative salient image regions and then computes the similarity based on these image regions in conjunction with the whole image. We use Spatial Transformer Networks (STN) for localizing salient regions. Extensive experiments on CUHK01 and CUHK03 datasets show that our method achieves the state-of-the-art performance.