Research map / Label-efficient and robust learning / Recognition and long-tail
SPUR: Scale-Partitioned Uncertainty Rectification for Robust UAV-on-UAV Interception
Chenqi Yan, Zhaoyu Zeng, Yifeng Yang, Jundong Zhou, Zhuoyuan Ni, Junqi Wu, Qinying Gu, Xinbing Wang, Nanyang Ye
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 group-wise minimax over scale partitions (ablation) | ERM YOLO-tiny · UAV-MultiView Far | mAP50: 90.98 -> 91.95 | Improved |
| add group-wise minimax over scale partitions (ablation) | ERM YOLO-tiny · UAV-MultiView overall | mAP50: 94.64 -> 95.51 | Improved |
| add uncertainty coupled into minimax weights (ablation) | Group-wise Minimax · UAV-MultiView Far | mAP50: 91.95 -> 86.76 | Hurt |
| add uncertainty coupled into minimax weights (ablation) | Group-wise Minimax · UAV-MultiView overall | mAP50: 95.51 -> 93.48 | Hurt |
| decouple adversary weights from uncertainty (proxy loss) (ablation) | Coupled Minimax+Uncertainty · UAV-MultiView Far | mAP50: 86.76 -> 92.18 | Improved |
| add decoupled uncertainty rectification (ablation) | Group-wise Minimax · Gaussian noise sigma=0.05 | mAP50: 90.84 -> 93.34 | Improved |
| add control-aware center penalty (ablation) | Decoupled Minimax+Uncertainty · UAV-MultiView overall | mAP50: 95.80 -> 96.10 | Improved |
| add control-aware center penalty (ablation) | Decoupled Minimax+Uncertainty · Gaussian noise sigma=0.05 | Center MAE: 3.3168 -> 3.3061 | Improved |
| add control-aware center penalty (ablation) | Decoupled Minimax+Uncertainty · UAV-MultiView Far | mAP50: 92.18 -> 92.75 | Improved |
| add group-wise minimax without uncertainty (ablation) | ERM YOLO-tiny · Gaussian noise sigma=0.05 | mAP50: 92.60 -> 90.84 | Hurt |
| add decoupled uncertainty rectification (ablation) | Group-wise Minimax · Motion blur k=5 | mAP50: 89.57 -> 91.43 | Improved |
| add coupled uncertainty to minimax (ablation) | Group-wise Minimax · Motion blur k=5 | Center MAE: 3.9072 -> 4.5565 | Hurt |
Closest papers by meaning
- AeroGS: Scale-Aware Gaussian Splatting for Pose-Free Dynamic UAV Scene Reconstruction (CVPR 2026 · similarity 0.84)
- PRISM: Priority-Guided Scanning in the Wavelet Domain for UAV Maritime Small Object Detection (NeurIPS 2026 · similarity 0.84)
- Learning Occlusion-Robust Vision Transformers for Real-Time UAV Tracking (CVPR 2025 · similarity 0.83)
- RemDet: Rethinking Efficient Model Design for UAV Object Detection (AAAI 2025 · similarity 0.83)
- M$^2$E-UAV: A Benchmark and Analysis for Onboard Motion-on-Motion Event-Based Tiny UAV Detection (NeurIPS 2026 · similarity 0.82)
- SOAR: Regression-based LiDAR Relocalization for UAVs (NeurIPS 2026 · similarity 0.82)
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.
