Memory Regulation and Alignment toward Generalizer RGB-Infrared Person
Feng Chen, Fei Wu, Qi Wu, Zhiguo Wan

TL;DR
This paper introduces MG-MRA, a novel memory regulation and alignment module that improves RGB-Infrared person re-identification by reducing domain shift effects through multi-granularity attribute incorporation and structural pattern matching.
Contribution
The paper proposes a new multi-granularity memory regulation and alignment module (MG-MRA) that enhances feature robustness and domain adaptation in RGB-Infrared person re-identification.
Findings
Outperforms state-of-the-art methods on RegDB and SYSU-MM01 datasets.
Effectively reduces over-confidence in discriminative features.
Improves cross-modality person re-identification accuracy.
Abstract
The domain shift, coming from unneglectable modality gap and non-overlapped identity classes between training and test sets, is a major issue of RGB-Infrared person re-identification. A key to tackle the inherent issue -- domain shift -- is to enforce the data distributions of the two domains to be similar. However, RGB-IR ReID always demands discriminative features, leading to over-rely feature sensitivity of seen classes, \textit{e.g.}, via attention-based feature alignment or metric learning. Therefore, predicting the unseen query category from predefined training classes may not be accurate and leads to a sub-optimal adversarial gradient. In this paper, we uncover it in a more explainable way and propose a novel multi-granularity memory regulation and alignment module (MG-MRA) to solve this issue. By explicitly incorporating a latent variable attribute, from fine-grained to coarse…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsVideo Surveillance and Tracking Methods · Human Pose and Action Recognition · Advanced Neural Network Applications
MethodsTest
