Meta Distribution Alignment for Generalizable Person Re-Identification

Abstract

Domain generalization person re-identification (DG-ReID) aims to train a model on source domains and generalize well on unseen domains. Vision Transformer usually yields better generalization ability than common CNN networks under distribution shifts. However, Transformerbased ReID models inevitably over-fit to domain-specific biases due to the supervised learning strategy on the source domain. We observe that while the global images of different IDs should have different features, their similar local parts (e.g., black backpack) are not bounded by this constraint. Motivated by this, we propose a pure Transformer model (termed Part-aware Transformer) for DG-ReID by designing a proxy task, named Cross-ID Similarity Learning (CSL), to mine local visual information shared by different IDs. This proxy task allows the model to learn generic features because it only cares about the visual similarity of the parts regardless of the ID labels, thus alleviating the side effect of domain-specific biases. Based on the local similarity obtained in CSL, a Part-guided Self-Distillation (PSD) is proposed to further improve the generalization of global features. Our method achieves stateof-the-art performance under most DG ReID settings. Under the Market→Duke setting, our method exceeds stateof-the-art by 10.9% and 12.8% in Rank1 and mAP, respectively. The code is available at https://github.com/ liyuke65535/Part-Aware-Transformer.

Transformer-based vehicle re-identification with multiple details

“…Its architecture, which relies on self-attention mechanisms, allows it to process entire sequences of data in parallel and capture long-range dependencies, making it highly efficient and scalable compared to traditional recurrent neural networks (RNNs) and convolutional neural networks (CNNs) [17,18]. At present, transformers [19,20] are being utilized across various computer vision tasks, garnering considerable interest from researchers for their robust modeling capabilities. In…”

Cross-Domain Person Re-Identification Based on Feature Fusion Invariance

“…The global features and local features are fused by a multiscale channel attention module (MSCAM) [16].…”

“…Li et al [15] introduced the concept of stream shape regularization, which achieves stream shape alignment and balanced distribution adaptation by weighting the distributional importance of the source and target domains and preserving the neighborhood structure. Moreover, a novel Meta Distribution Alignment (MDA) method [16] is proposed to enable them to share similar distribution in a test-time training fashion. Ni et al [17] propose a pure Transformer model (termed Part-aware Transformer) for DG-ReID by designing a proxy task to mine local visual information shared by different IDs.…”

“…Moreover, a novel Meta Distribution Alignment (MDA) method [16] is proposed to enable them to share similar distribution in a test-time training fashion. Ni et al [17] propose a pure Transformer model (termed Part-aware Transformer) for DG-ReID by designing a proxy task to mine local visual information shared by different IDs. Jiao et al [18] develop a new normalization scheme called Dynamically Transformed Instance Normalization (DTIN) to alleviate the drawback of IN (i.e., the limitation of eliminating discriminative patterns).…”

Robust Ensemble Person Re-Identification Via Orthogonal Fusion with Occlusion Handling

“…Group-aware label transfer (GLT) (20) was proposed for unsupervised domain adaptive person Re-ID; a similar unsupervised person Re-ID method was presented in (21) that focuses on considering the distribution discrepancy between cameras. A meta-distribution alignment strategy has been proposed for domain generalizable person ReID (22). Inspired by the latest advances in transformers, a new part-aware transformer (PAT) was developed for the occluded person Re-ID in (23) through a transformer encoder-decoder architecture.…”