arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2603.12912cs.CVcs.AI
收藏

FedBPrompt: 通过身体分布感知的视觉提示实现联邦领域泛化的人重识别

FedBPrompt: Federated Domain Generalization Person Re-Identification via Body Distribution Aware Visual Prompts

School of Computer Science and Technology, Wuhan University of Science and Technology, China(武汉理工大学计算机科学与技术学院,中国);NERC for Multimedia Software, School of Computer Science, Wuhan University, China(武汉大学计算机学院,中国);Nanyang Technological University, Singapore(新加坡南洋理工大学);Changsha Bus Group, China(中国长沙公交集团);Central South University of Forestry and Technology, China(中国林业科技大学);Harbin Institute of Technology Zhengzhou Research Institute, China(哈尔滨工业大学郑州研究院,中国)

Xin Xu, Weilong Li, Wei Liu, Wenke Huang, Zhixi Yu, Bin Yang, Xiaoying Liao, Kui Jiang

AI总结:

本文提出FedBPrompt,通过引入可学习的视觉提示引导Transformer关注行人区域,解决联邦领域泛化重识别中域偏移问题,同时设计轻量级提示微调策略降低通信开销。

摘要:

联邦领域泛化的人重识别(FedDG-ReID)通过从分布式数据中学习域不变表示。尽管视觉Transformer(ViT)被广泛采用,但其全局注意力常难以区分高相似背景或多样视角的行人——这一挑战在FedDG-ReID中因跨客户端分布偏移而加剧。为解决此问题,我们提出联邦身体分布感知视觉提示(FedBPrompt),引入可学习视觉提示引导Transformer注意力聚焦于行人区域。FedBPrompt采用包含整体全身体提示和身体部分对齐提示的Body Distribution Aware Visual Prompts Mechanism(BAPM),以抑制跨客户端背景噪声并捕捉鲁棒于姿态和视角变化的细粒度细节。为缓解高通信成本,我们设计了基于提示的微调策略(PFTS),冻结ViT主干并仅更新轻量级提示,显著降低通信开销同时保持适应性。大量实验表明,BAPM有效提升特征判别性和跨域泛化能力,而PFTS在仅几次聚合轮次内实现显著性能提升。此外,BAPM和PFTS可轻松集成到现有ViT-based FedDG-ReID框架中,使FedBPrompt成为联邦人重识别的灵活有效解决方案。代码可在https://github.com/leavlong/FedBPrompt获取。

英文摘要:

Federated Domain Generalization for Person Re-Identification (FedDG-ReID) learns domain-invariant representations from decentralized data. While Vision Transformer (ViT) is widely adopted, its global attention often fails to distinguish pedestrians from high similarity backgrounds or diverse viewpoints -- a challenge amplified by cross-client distribution shifts in FedDG-ReID. To address this, we propose Federated Body Distribution Aware Visual Prompt (FedBPrompt), introducing learnable visual prompts to guide Transformer attention toward pedestrian-centric regions. FedBPrompt employs a Body Distribution Aware Visual Prompts Mechanism (BAPM) comprising: Holistic Full Body Prompts to suppress cross-client background noise, and Body Part Alignment Prompts to capture fine-grained details robust to pose and viewpoint variations. To mitigate high communication costs, we design a Prompt-based Fine-Tuning Strategy (PFTS) that freezes the ViT backbone and updates only lightweight prompts, significantly reducing communication overhead while maintaining adaptability. Extensive experiments demonstrate that BAPM effectively enhances feature discrimination and cross-domain generalization, while PFTS achieves notable performance gains within only a few aggregation rounds. Moreover, both BAPM and PFTS can be easily integrated into existing ViT-based FedDG-ReID frameworks, making FedBPrompt a flexible and effective solution for federated person re-identification. The code is available at https://github.com/leavlong/FedBPrompt.

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。

↑