# Person re-identification with fusion of hand-crafted and deep pose-based body region features

## TL;DR

This paper introduces a fusion approach combining handcrafted and deep pose-based features for person re-identification, improving accuracy especially on unseen individuals by addressing misalignment and pose variations.

## Contribution

It proposes a novel fusion of handcrafted and deep features guided by pose information, enhancing re-ID performance on unseen test images.

## Key findings

- Achieves state-of-the-art results on VIPer, DukeMTMC-reID, Market-1501, CUHK03 datasets.
- Effectively handles pose variations and misalignments.
- Improves generalization to unseen persons.

## Abstract

Person re-identification (re-ID) aims to accurately re- trieve a person from a large-scale database of images cap- tured across multiple cameras. Existing works learn deep representations using a large training subset of unique per- sons. However, identifying unseen persons is critical for a good re-ID algorithm. Moreover, the misalignment be- tween person crops to detection errors or pose variations leads to poor feature matching. In this work, we present a fusion of handcrafted features and deep feature representa- tion learned using multiple body parts to complement the global body features that achieves high performance on un- seen test images. Pose information is used to detect body regions that are passed through Convolutional Neural Net- works (CNN) to guide feature learning. Finally, a metric learning step enables robust distance matching on a dis- criminative subspace. Experimental results on 4 popular re-ID benchmark datasets namely VIPer, DukeMTMC-reID, Market-1501 and CUHK03 show that the proposed method achieves state-of-the-art performance in image-based per- son re-identification.