提出首个端到端行人搜索预训练框架，提升模型鲁棒性与性能。
Swap Path Network for Robust Person Search Pre-training
- 设计可切换的双路径网络，兼顾查询与目标中心训练。
- 在CUHK-SYSU上达96.4% mAP，PRW上达61.2% mAP，刷新纪录。
- 仅用弱标注框即可训练，适合噪声数据场景下的预训练。
行人搜索旨在从画廊场景中检测并排序与查询图像匹配的行人。现有模型通常采用特征提取主干网络，并分别配置检测与重识别分支。尽管视觉主干的预训练方法已成熟，但针对行人搜索任务的附加模块预训练尚未被探索。本文首次提出端到端行人搜索预训练框架，将任务拆分为以对象为中心和以查询为中心两种范式。结果显示，以查询为中心的方法对标签噪声更鲁棒，且仅需弱标注的行人边界框即可训练。我们进一步提出新型模型Swap Path Net（SPNet），可共享权重在两种训练目标间切换。实验表明，采用查询中心预训练+对象中心微调的策略，在标准的PRW与CUHK-SYSU数据集上均达到当前最优性能：在CUHK-SYSU上获得96.4% mAP，PRW上达61.2% mAP。此外，该方法在效率、效果和鲁棒性方面均优于近期仅预训练主干的替代方案。
原文摘要 · Abstract (English)
In person search, we detect and rank matches to a query person image within a set of gallery scenes. Most person search models make use of a feature extraction backbone, followed by separate heads for detection and re-identification. While pre-training methods for vision backbones are well-established, pre-training additional modules for the person search task has not been previously examined. In this work, we present the first framework for end-to-end person search pre-training. Our framework splits person search into object-centric and query-centric methodologies, and we show that the query-centric framing is robust to label noise, and trainable using only weakly-labeled person bounding boxes. Further, we provide a novel model dubbed Swap Path Net (SPNet) which implements both query-centric and object-centric training objectives, and can swap between the two while using the same weights. Using SPNet, we show that query-centric pre-training, followed by object-centric fine-tuning, achieves state-of-the-art results on the standard PRW and CUHK-SYSU person search benchmarks, with 96.4% mAP on CUHK-SYSU and 61.2% mAP on PRW. In addition, we show that our method is more effective, efficient, and robust for person search pre-training than recent backbone-only pre-training alternatives.
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。