Research map / Detection and segmentation / Tracking
Rethinking Occlusion Modeling for UAV Tracking
Jian Zhang, Xincheng Yu, Yi Lin
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Clustered Occlusion Modeling (COM) training masks (ablation) | OCTrack-DeiT · UAVDT | precision: 80.4 -> 85.2 | Improved |
| add COM training masks (ablation) | OCTrack-ViT · UAVDT | AUC: 58.8 -> 62.0 | Improved |
| add Cost-Aware Depth Bias (CADB) on top of COM (ablation) | OCTrack-DeiT · UAVDT | FPS: 197.6 -> 265.2 | Improved |
| add CADB on top of COM (ablation) | OCTrack-DeiT · UAVDT | precision: 85.2 -> 85.0 | No clear effect |
| replace random masking with clustered masking (COM) (ablation) | OCTrack-DeiT · VisDrone2018 | AUC: 62.9 -> 67.0 | Improved |
| replace Cox process masking with COM (ablation) | OCTrack-DeiT · VisDrone2018 | AUC: 65.9 -> 67.0 | Improved |
| use CutMix or AdAutoMix instead of COM (ablation) | OCTrack-DeiT · VisDrone2018 | AUC: 67.0 -> 63.6 (CutMix), 63.2 (AdAutoMix) | Hurt |
| increase CADB bias strength kappa from 0.0 to 0.5 (ablation) | OCTrack-DeiT · UAV123 | FPS: 197.6 -> 278.4 | Improved |
| increase CADB bias strength kappa from 0.0 to 0.5 (ablation) | OCTrack-DeiT · UAV123 | AUC: 66.9 -> 66.0 | No clear effect |
| use dense occlusion style instead of balanced (ablation) | OCTrack-DeiT · UAV123 | AUC: 66.9 -> 67.7 | Improved |
| use sparse or dense occlusion style instead of balanced (ablation) | OCTrack-DeiT · UAVDT | precision: 85.0 -> 81.8 (dense), 84.2 (sparse) | Hurt |
Closest papers by meaning
- Learning Occlusion-Robust Vision Transformers for Real-Time UAV Tracking (CVPR 2025 · similarity 0.92)
- Learning Adaptive and View-Invariant Vision Transformer for Real-Time UAV Tracking (ICML 2024 · similarity 0.86)
- MUST: The First Dataset and Unified Framework for Multispectral UAV Single Object Tracking (CVPR 2025 · similarity 0.85)
- OAMaskFlow: Occlusion-Aware Motion Mask for Scene Flow (AAAI 2025 · similarity 0.84)
- UAV3D: A Large-scale 3D Perception Benchmark for Unmanned Aerial Vehicles (NeurIPS 2024 · similarity 0.84)
- CorrBEV: Multi-View 3D Object Detection by Correlation Learning with Multi-modal Prototypes (CVPR 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
