Person Re-identification based on Robust Features in Open-world

TL;DR

This paper introduces a robust person re-identification method for open-world scenarios, utilizing pose estimation and dynamic time warping to handle multi-factor variability and improve accuracy over existing models.

Contribution

The paper proposes a novel re-ID approach combining pose estimation with DTW and introduces a new dataset reflecting real-world multi-factor variability.

Findings

Achieves Rank-1 accuracy of 60.9% on the new dataset.

Outperforms most existing re-ID models in experiments.

Effectively handles multi-factor variables like clothing and modality.

Abstract

Deep learning technology promotes the rapid development of person re-identifica-tion (re-ID). However, some challenges are still existing in the open-world. First, the existing re-ID research usually assumes only one factor variable (view, clothing, pedestrian pose, pedestrian occlusion, image resolution, RGB/IR modality) changing, ignoring the complexity of multi-factor variables in the open-world. Second, the existing re-ID methods are over depend on clothing color and other apparent features of pedestrian, which are easily disguised or changed. In addition, the lack of benchmark datasets containing multi-factor variables is also hindering the practically application of re-ID in the open-world. In this paper, we propose a low-cost and high-efficiency method to solve shortcomings of the existing re-ID research, such as unreliable feature selection, low efficiency of feature extraction, single research variable, etc. Our approach based on pose estimation model improved by group convolution to obtain the continuous key points of pedestrian, and utilize dynamic time warping (DTW) to measure the similarity of features between different pedestrians. At the same time, to verify the effectiveness of our method, we provide a miniature dataset which is closer to the real world and includes pedestrian changing clothes and cross-modality factor variables fusion. Extensive experiments are conducted and the results show that our method achieves Rank-1: 60.9%, Rank-5: 78.1%, and mAP: 49.2% on this dataset, which exceeds most existing state-of-art re-ID models.