UMDATrack：恶劣天气条件下的统一多域自适应跟踪
UMDATrack: Unified Multi-Domain Adaptive Tracking Under Adverse Weather Conditions
- Beijing University of Posts and Telecommunications（北京邮电大学）
- Sun Yat-sen University（中山大学）
- University of Chinese Academy of Sciences（中国科学院大学）
- MoE Key Laboratory of Information Technology（教育部信息科学技术重点实验室）
- Guangdong Key Laboratory of Information Security Technology（广东省信息安全技术重点实验室）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本文提出UMDATrack，通过可控场景生成器合成少量多天气无标签视频，结合域定制适配器和目标感知置信度对齐模块，在统一域自适应框架下实现恶劣天气条件下的高性能视觉目标跟踪。
AI中文摘要：
视觉目标跟踪在过去几十年中取得了可喜的进展。现有方法大多侧重于在光照良好的日间数据中学习目标表示，而在夜间或雾天等恶劣天气条件下的无约束真实场景中，巨大的域偏移会导致性能显著下降。本文提出UMDATrack，能够在统一的域自适应框架内，在多种恶劣天气条件下保持高质量的目标状态预测。具体而言，我们首先使用一个可控场景生成器，在不同文本提示的引导下，合成少量无标签视频（少于源日间数据集中2%的帧），覆盖多种天气条件。随后，我们设计了一个简单而有效的域定制适配器（DCA），使目标对象的表示能够快速适应各种天气条件，而无需冗余的模型更新。此外，为了增强源域与目标域之间的定位一致性，我们基于最优传输定理提出了目标感知置信度对齐模块（TCA）。大量实验表明，UMDATrack能够超越现有先进视觉跟踪器，并以显著优势取得新的最先进性能。代码可在https://github.com/Z-Z188/UMDATrack获取。
英文摘要：
Visual object tracking has gained promising progress in past decades. Most of the existing approaches focus on learning target representation in well-conditioned daytime data, while for the unconstrained real-world scenarios with adverse weather conditions, e.g. nighttime or foggy environment, the tremendous domain shift leads to significant performance degradation. In this paper, we propose UMDATrack, which is capable of maintaining high-quality target state prediction under various adverse weather conditions within a unified domain adaptation framework. Specifically, we first use a controllable scenario generator to synthesize a small amount of unlabeled videos (less than 2% frames in source daytime datasets) in multiple weather conditions under the guidance of different text prompts. Afterwards, we design a simple yet effective domain-customized adapter (DCA), allowing the target objects' representation to rapidly adapt to various weather conditions without redundant model updating. Furthermore, to enhance the localization consistency between source and target domains, we propose a target-aware confidence alignment module (TCA) following optimal transport theorem. Extensive experiments demonstrate that UMDATrack can surpass existing advanced visual trackers and lead new state-of-the-art performance by a significant margin. Our code is available at https://github.com/Z-Z188/UMDATrack.
