发表机构
Trinity College Dublin; School of Zoology, Trinity College Dublin; Sigmedia Group, Electronic and Electrical Engineering, Trinity College Dublin（都柏林圣三一学院; 都柏林圣三一学院动物学学院; 都柏林圣三一学院电子与电气工程系Sigmedia团队）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对全景相机中小而快速移动目标的跟踪难题，重新工程化SORT算法，提出接缝感知运动模型和关联代价，并构建OmniSmall基准，显著提升跟踪性能且保持CPU-only。
AI 中文摘要
多目标跟踪（MOT）在城市监控和自动驾驶领域已取得快速发展，然而许多跟踪器依赖基于ReID和Transformer的外观编码器，并针对标准视场角（FoV）相机设计。这些假设在低成本全景部署中失效，因为等距柱状投影引入了接缝不连续性，且目标表现为小而快速移动。我们解决了使用全景相机在偏远环境中捕获的飞行动物的多目标跟踪问题。我们提出了一个轻量级框架，针对这种几何结构重新工程化基于SORT的跟踪，包括（i）一种接缝感知运动模型，使卡尔曼状态在接缝处保持连续，（ii）一种复合的接缝感知关联代价，将包裹的欧几里得项与GIoU配对，以及（iii）OmniSmall，一个全新的全景野生动物视频基准。在我们的新数据集上，使用真实标注检测时，我们的修改相比OCSORT在HOTA上提高了+8.51，在MOTA上提高了+9.41，在IDF1上提高了+10.17；使用YOLOX检测时，增益缩小至HOTA +1.95。我们提出的方法在OmniSmall上提升了跟踪性能，并在JRDB上保持竞争力，同时无需添加外观编码器，且跟踪阶段仅使用CPU。我们的数据集和源代码可在以下网址获取：this https URL。
英文摘要
Multi-object tracking (MOT) has advanced rapidly in urban surveillance and autonomous driving, yet many trackers rely on ReID- and transformer-based appearance encoders and are designed for standard FoV cameras. These assumptions break down for low-cost omnidirectional deployments, where equirectangular projection introduces seam discontinuities and targets appear to be small and fast-moving. We address multi-object tracking of flying animals captured in remote environments using omnidirectional cameras. We propose a lightweight framework that re-engineers SORT-based tracking for this geometry, including (i) a Seam-Aware Motion Model that keeps the Kalman state continuous across the seam, (ii) a composite seam-aware association cost that pairs a wrapped Euclidean term with GIoU, and (iii) OmniSmall, a new benchmark of omnidirectional wildlife footage. On our new dataset, with ground-truth detections, our modifications improved over OCSORT by +8.51 HOTA, +9.41 MOTA, and +10.17 IDF1; with YOLOX detections the gain narrows to +1.95 HOTA. Our proposed methods improved tracking performance on OmniSmall and remained competitive on JRDB without adding appearance encoders while keeping the tracking stage CPU-only. Our dataset and source code are available at: https://github.com/Xin-Shu/OmniSORT.git.
CommentsAccepted by IEEE 28th International Workshop on Multimedia Signal Processing (MMSP)
