# Improving Person Re-identification by Attribute and Identity Learning

## TL;DR

This paper introduces an attribute-person recognition (APR) network that jointly learns person re-identification and attribute recognition, leveraging attribute labels to enhance discriminative feature learning and improve re-ID accuracy.

## Contribution

The paper proposes a multi-task APR network that combines identity and attribute learning, with systematic annotation and analysis of attribute benefits for re-ID.

## Key findings

- APR achieves competitive re-ID performance on large datasets.
- Using attribute information speeds up retrieval by ten times.
- APR improves attribute recognition accuracy.

## Abstract

Person re-identification (re-ID) and attribute recognition share a common target at learning pedestrian descriptions. Their difference consists in the granularity. Most existing re-ID methods only take identity labels of pedestrians into consideration. However, we find the attributes, containing detailed local descriptions, are beneficial in allowing the re-ID model to learn more discriminative feature representations. In this paper, based on the complementarity of attribute labels and ID labels, we propose an attribute-person recognition (APR) network, a multi-task network which learns a re-ID embedding and at the same time predicts pedestrian attributes. We manually annotate attribute labels for two large-scale re-ID datasets, and systematically investigate how person re-ID and attribute recognition benefit from each other. In addition, we re-weight the attribute predictions considering the dependencies and correlations among the attributes. The experimental results on two large-scale re-ID benchmarks demonstrate that by learning a more discriminative representation, APR achieves competitive re-ID performance compared with the state-of-the-art methods. We use APR to speed up the retrieval process by ten times with a minor accuracy drop of 2.92% on Market-1501. Besides, we also apply APR on the attribute recognition task and demonstrate improvement over the baselines.