arXiv cs.AI· Jiangong Xiao, Zhe Sun, Kanzhong Yao, Yuanbo Bi, Haofei Zhao, Ruixuan Hu, Guan Huang, Xuelong Li·· 9 天前AI 评分17
跨水空界面的连续真值构建与恢复策略：CMRP 提升 MixFormerV2 跟踪性能
Continuous Ground-Truth Construction and a Recovery Policy for Air--Water Robotic Tracking
AI 导读
针对跨水空界面视觉跟踪中飞溅、气泡、折射等导致观测失效的问题，研究者构建了同步相机帧与动捕位姿的真值流水线，生成 22,346 帧的跨介质评测集。其提出的 Cross-Medium Recovery Policy（CMRP）以置信度触发模板选择，为 MixFormerV2 提供固定初始模板、窗口最优预触发模板和触发帧 Kalman 引导裁剪，无需重训练视觉骨干。
正文 · 原文
Abstract:Visual tracking across the air-water interface is challenged by splashes, bubbles, refraction, reflections, and abrupt appearance changes that can temporarily invalidate observations. This setting poses two coupled difficulties: first, for evaluation, image-only annotation cannot reliably describe the target's physical location during visual blindness; second, for online tracking, corrupted observations can contaminate motion estimates and appearance templates. We address the first difficulty with a construction pipeline that synchronizes camera frames with motion-capture poses, projects known target geometry, corrects underwater projection with a medium-gated residual, and subjects the annotations to manual review. This yields an evaluation-only cross-medium test set of 22,346 frames. We further introduce a Cross-Medium Recovery Policy (CMRP) centered on confidence-triggered template selection. It supplies MixFormerV2 with the fixed initial template, a window-best pre-trigger template, and a trigger-frame Kalman-guided image crop, together with their associated weights, without retraining the visual backbone. In the accuracy evaluation, CMRP achieves 49.90 Macro Success AUC, 2.95 points above MixFormerV2 Official. On selected cross-medium transition and occlusion-recovery intervals, CMRP increases MixFormerV2 tracking coverage from 47.91\% to 50.43\% relative to Official updating, while mean loss-to-recovery latency over successfully recovered videos decreases from 55.3 to 49.3 frames.
| Comments: | 8 pages, 6 figures. Submitted to IEEE International Conference on Robotics and Automation (ICRA 2027) |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.11096 [cs.CV] |
| (or arXiv:2610.11096v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.11096 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jiangong Xiao [view email]
[v1]
Thu, 8 Oct 2026 02:11:43 UTC (13,195 KB)
来源：arXiv cs.AI · arxiv.org