发表机构
Harbin Institute of Technology, Shenzhen; Great Bay University, Dongguan; Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences（哈尔滨工业大学（深圳）; 大湾区大学（东莞）; 中国科学院深圳先进技术研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对恶劣环境下运动目标感知难题，提出RGBTR-Motion基准和SAM-Radar框架，融合雷达、RGB和热成像实现鲁棒分割与跟踪，显著提升性能。
AI 中文摘要
运动目标感知必须决定哪些图像区域对应真实运动，并随时间保持每个实例的身份识别。从外观、光流或估计轨迹中读取运动的方法，在光照不足、恶劣天气、反射和遮挡条件下会丢失这些证据。雷达是一种自然的解决方案，因为它直接测量径向速度，而不是从光度对应关系中推断。然而，现有的基准测试并未同时为监控场景提供雷达测量、密集运动实例掩码和时间一致的身份信息。因此，我们引入了RGBTR-Motion，一个同步且校准的固定相机基准，它将RGB、热成像和雷达流与密集实例掩码以及跨多种监控场景的时间一致身份配对。我们还开发了SAM-Radar，一个基于SAM 3的RGB、热成像和雷达分割与跟踪框架。SAM-Radar的雷达感知检测器将校准的RGBT特征与投影到其图像位置的雷达回波融合，运动监督（实现为这些投影回波的前景分类）教会检测器无需任何文本提示即可拒绝杂波。跟踪器将接受的雷达回波与单个轨迹关联，并将其作为视觉退化目标仍然存在的物理证据。这使其能够桥接短暂的能见度低或遮挡时期，并将重新出现的目标重新连接到其现有身份，而不是开始新的轨迹。SAM-Radar达到了0.7027的IoU和0.8090的F1-50，并将MOTA、HOTA和IDF1分别比最强竞争值提高了0.2977、0.1603和0.2857。
英文摘要
Moving-object perception must decide which image regions correspond to real motion and keep every instance identified over time. Methods that read motion from appearance, optical flow, or estimated trajectories lose that evidence under poor illumination, adverse weather, reflections, and occlusion. Radar is a natural remedy because it measures radial velocity directly instead of inferring it from photometric correspondence. However, existing benchmarks do not jointly provide radar measurements, dense moving-instance masks, and temporally consistent identities for surveillance. We therefore introduce RGBTR-Motion, a synchronized and calibrated fixed-camera benchmark that pairs RGB, thermal, and radar streams with dense instance masks and temporally consistent identities across diverse surveillance scenes. We also develop SAM-Radar, an RGB, thermal, and radar-based segmentation and tracking framework built on SAM 3. SAM-Radar's radar-aware detector fuses calibrated RGBT features with radar returns that are grounded at their projected image locations, and motion supervision, implemented as foreground classification of those projected returns, teaches the detector to reject clutter without any text prompt. The tracker associates accepted radar returns with individual trajectories and uses them as physical evidence that a visually degraded target remains present. This allows it to bridge short periods of low visibility or occlusion and reconnect a reappearing target to its existing identity instead of starting a new track. SAM-Radar attains 0.7027 IoU and 0.8090 F1-50, and raises MOTA, HOTA, and IDF1 by 0.2977, 0.1603, and 0.2857 over the strongest competing values.
Comments9 pages, 5figures
