AAformer：用于行人重识别的自动对齐Transformer
AAformer: Auto-Aligned Transformer for Person Re-Identification
AI总结：
本文提出AAformer，首次在Transformer中引入自动对齐机制，通过可学习的部件标记和最优传输算法，在补丁级别自动定位人体与非人体部位，提取部件级特征用于行人重识别，实验验证其优于现有方法。
摘要：
在行人重识别（re-ID）中，从行人图像中提取部件级特征已被验证对于提供细粒度信息至关重要。现有的大多数基于CNN的方法仅能粗略定位人体部位，或依赖预训练的人体解析模型，且无法定位可识别的非人体部位（如背包）。本文首次在Transformer架构中引入对齐方案，并提出自动对齐Transformer（AAformer），以在补丁级别自动定位人体部位和非人体部位。我们引入了“部件标记（[PART]s）”，这些是可学习的向量，用于在Transformer中提取部件特征。一个[PART]仅与自注意力中的局部补丁子集交互，并学习成为部件表示。为了自适应地将图像补丁分组到不同的子集中，我们设计了自动对齐。自动对齐采用最优传输（OT）算法的快速变体，以[PART]s为原型，在线将补丁嵌入聚类为若干组。AAformer将部件对齐集成到自注意力中，输出的[PART]s可直接用作检索的部件特征。大量实验验证了[PART]s的有效性以及AAformer相对于各种最先进方法的优越性。
英文摘要：
In person re-identification (re-ID), extracting part-level features from person images has been verified to be crucial to offer fine-grained information. Most of the existing CNN-based methods only locate the human parts coarsely, or rely on pretrained human parsing models and fail in locating the identifiable nonhuman parts (e.g., knapsack). In this article, we introduce an alignment scheme in transformer architecture for the first time and propose the auto-aligned transformer (AAformer) to automatically locate both the human parts and nonhuman ones at patch level. We introduce the "Part tokens ([PART]s)", which are learnable vectors, to extract part features in the transformer. A [PART] only interacts with a local subset of patches in self-attention and learns to be the part representation. To adaptively group the image patches into different subsets, we design the auto-alignment. Auto-alignment employs a fast variant of optimal transport (OT) algorithm to online cluster the patch embeddings into several groups with the [PART]s as their prototypes. AAformer integrates the part alignment into the self-attention and the output [PART]s can be directly used as part features for retrieval. Extensive experiments validate the effectiveness of [PART]s and the superiority of AAformer over various state-of-the-art methods.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
