提出身份感知特征解耦方法，提升换装场景下的人体重识别精度。
Identity-aware Feature Decoupling Learning for Clothing-change Person Re-identification
- 双流架构：主干流+注意力流，通过衣物掩码图生成身份注意力权重。
- 在多个主流换装重识别数据集上超越现有模型，显著提升识别准确率。
- 设计去衣物偏置模块，减少衣物特征干扰，增强身份相关区域表征。
换装人体重识别（CC Re-ID）因应用前景广阔而受到广泛关注。现有方法难以从原始RGB图像中充分提取身份相关信息。本文提出一种身份感知特征解耦（IFD）学习框架，以挖掘身份相关特征。IFD采用双流结构，包括主干流和注意力流。注意力流以衣物掩码图像为输入，生成身份注意力权重，将空间知识有效传递至主干流，并突出富含身份信息的区域。为消除两流输入间的语义差距，本文在主干流中引入专用的去衣物偏置模块，对衣物相关区域特征进行正则化。大量实验结果表明，该框架在多个广泛使用的CC Re-ID数据集上优于其他基线模型。
原文摘要 · Abstract (English)
Clothing-change person re-identification (CC Re-ID) has attracted increasing attention in recent years due to its application prospect. Most existing works struggle to adequately extract the ID-related information from the original RGB images. In this paper, we propose an Identity-aware Feature Decoupling (IFD) learning framework to mine identity-related features. Particularly, IFD exploits a dual stream architecture that consists of a main stream and an attention stream. The attention stream takes the clothing-masked images as inputs and derives the identity attention weights for effectively transferring the spatial knowledge to the main stream and highlighting the regions with abundant identity-related information. To eliminate the semantic gap between the inputs of two streams, we propose a clothing bias diminishing module specific to the main stream to regularize the features of clothing-relevant regions. Extensive experimental results demonstrate that our framework outperforms other baseline models on several widely-used CC Re-ID datasets.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。