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 2405.02832cs.CV
收藏

快速单阶段无监督域自适应行人搜索

Fast One-Stage Unsupervised Domain Adaptive Person Search

The College of Information and Science Technology, Dalian Maritime University(大连海事大学信息科学技术学院);The College School of Cyber Security and Computer, Hebei University(河北大学网络空间安全与计算机学院);The College of Computer Science and Artificial Intelligence, Wenzhou University(温州大学计算机科学与人工智能学院);Key Laboratory of Knowledge Engineering with Big Data, Hefei University of Technology(合肥工业大学大数据知识工程重点实验室)

Tianxiang Cui, Huibing Wang, Jinjia Peng, Ruoxi Deng, Xianping Fu, Yang Wang

AI总结:

本文提出快速单阶段无监督行人搜索方法FOUS,通过端到端集成域对齐与原型引导标签自适应,避免迭代聚类,在CUHK-SYSU和PRW上达到SOTA性能。

摘要:

无监督行人搜索旨在在没有标注的情况下,从场景图像库中定位特定的目标行人,由于未标注域的意外变化,这一任务极具挑战性。然而,现有大多数方法致力于开发多阶段模型来适应域变化,同时使用聚类进行迭代模型训练,这不可避免地增加了模型复杂度。为了解决这一问题,我们提出了一种快速单阶段无监督行人搜索方法(FOUS),它以端到端的方式将域自适应与标签自适应互补地集成在一起,无需迭代聚类。为了最小化域差异,FOUS引入了一个基于注意力的域对齐模块(ADAM),该模块不仅能够为检测和ReID任务对齐各种域,还能构建注意力机制以减少无监督检测产生的低质量候选对象的不利影响。此外,为了避免冗余的迭代聚类模式,FOUS采用了一种原型引导的标注方法,该方法最小化了部分样本的冗余相关性计算,并高效地分配带有噪声的粗标签组。这些粗标签组将通过具有自适应选择策略的标签灵活训练网络不断细化。借助自适应后的域和标签,FOUS在两个基准数据集CUHK-SYSU和PRW上取得了最先进(SOTA)的性能。代码可在 https://github.com/whbdmu/FOUS 获取。

英文摘要:

Unsupervised person search aims to localize a particular target person from a gallery set of scene images without annotations, which is extremely challenging due to the unexpected variations of the unlabeled domains. However, most existing methods dedicate to developing multi-stage models to adapt domain variations while using clustering for iterative model training, which inevitably increases model complexity. To address this issue, we propose a Fast One-stage Unsupervised person Search (FOUS) which complementary integrates domain adaptaion with label adaptaion within an end-to-end manner without iterative clustering. To minimize the domain discrepancy, FOUS introduced an Attention-based Domain Alignment Module (ADAM) which can not only align various domains for both detection and ReID tasks but also construct an attention mechanism to reduce the adverse impacts of low-quality candidates resulting from unsupervised detection. Moreover, to avoid the redundant iterative clustering mode, FOUS adopts a prototype-guided labeling method which minimizes redundant correlation computations for partial samples and assigns noisy coarse label groups efficiently. The coarse label groups will be continuously refined via label-flexible training network with an adaptive selection strategy. With the adapted domains and labels, FOUS can achieve the state-of-the-art (SOTA) performance on two benchmark datasets, CUHK-SYSU and PRW. The code is available at https://github.com/whbdmu/FOUS.

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。

↑