Memory Regulation and Alignment toward Generalizer RGB-Infrared Person

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, textite.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 semantic granularity, into intermediate features, our method could alleviate the over-confidence of the model about discriminative features of seen classes. Moreover, instead of matching discriminative features by traversing nearest neighbor, sparse attributes, textiti.e., global structural pattern, are recollected with respect to features and assigned to measure pair-wise image similarity in hashing. Extensive experiments on RegDB citeRegDB and SYSU-MM01 citeSYSU show the superiority of the proposed method that outperforms existing state-of-the-art methods. Our code is available in https://github.com/Chenfeng1271/MGMRA.