Dual-Stream IB-Enhanced Framework for Video-Based Person Re-Identification
Authors:
- Xu WangXWXu Wang
- Yaqiao LiaoYLYaqiao Liao
- Guowen KuangGKGuowen Kuang
- Jinfeng YangJYJinfeng Yang
Open Access
- 1 April 2025
- journal article
- Published by Tsinghua University Press in CAAI Artificial Intelligence Research
- Vol. 4
Abstract
Person re-identification (ReID) is a desirable yet challenging issue in computer vision, which has numerous potential applications in the public security area. In this paper, the information bottleneck (IB) theory is introduced to filter out the undiscriminating information and noise, and keep the distinctive information for persons, to identify the assigned person from the candidates. To meet the requirement of the IB theory, a dual-stream IB-enhanced (DSIB) ReID framework is proposed to make unentangled estimations of the expectation and variance of the feature map. Moreover, the unconditional feature distribution is relaxed from standard normal distribution to Gaussian distribution, to enrich the information contained in each feature channel. Comprehensive experiments are conducted with several baseline models on datasets MARS, LS-VID, iLiDS-VID, and PRID-2011. The results reveal that our proposed DSIB framework has improved the performance of all baseline models, and it outperforms most state-of-the-art methods in terms of mean average precision (mAP) and Rank-1 precision.Keywords
This publication has 49 references indexed in Scilit:
- Cross-Platform Video Person ReID: A New Benchmark Dataset and Adaptation Approach2024
- Video-Based Person Re-Identification with Long Short-Term Representation Learning2023
- ResT-ReID: Transformer block-based residual learning for person re-identificationPattern Recognition Letters, 2022
- Temporal Complementary Learning for Video Person Re-identification2020
- Temporal Coherence or Temporal Motion: Which Is More Critical for Video-Based Person Re-identification?2020
- Appearance-Preserving 3D Convolution for Video-Based Person Re-identification2020
- MARS: A Video Benchmark for Large-Scale Person Re-Identification2016
- Temporal Segment Networks: Towards Good Practices for Deep Action Recognition2016
- Person Re-identification by Video Ranking2014
- Person Re-identification by Descriptive and Discriminative Classification2011