Research map / Architectures and efficiency / RNNs and normalization
UAST: Unified Active Search and Tracking for Arbitrary Targets with UAVs
Liang Qin, Min Wang, Xingyu Lu, Aowen Qiu, Wengang Zhou, Houqiang Li
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 search mechanism (ablation) | UAST · Long-term tracking 5 m/s | success rate: 0.52 -> 0.89 | Improved |
| add tracking-aware visibility loss (ablation) | UAST · FoV center distance | center distance (m): 0.82 -> 0.57 | Improved |
| add tracking-aware visibility loss (ablation) | UAST · Long-term tracking 5 m/s | success rate: 0.82 -> 0.89 | Improved |
| add data construction strategy (ablation) | UAST · Short-range tracking 5 m/s | success rate: 0.80 -> 0.93 | Improved |
| add rule-based point search (guidance point) policy (ablation) | UAST · Long-term tracking 5 m/s | success rate: 0.33 -> 0.89 | Improved |
| add search mechanism (ablation) | UAST · Short-range tracking 5 m/s | success rate: 0.88 -> 0.93 | Improved |
| train Yopov2 on multiple target versions (ablation) | Yopov2 · Short-range 6 m/s | success rate: 0.80 -> 0.60 | Hurt |
| use UAST rule-based search instead of frontier exploration (RACER, FALCON) | UAST · Active search | search time (s): 176.62 -> 54.49 | Improved |
| use UAST instead of mapping-based trackers at high speed | UAST · Short-range 5 m/s | success rate: 0.15 -> 0.93 | Improved |
| UAST vs Yopov2 cycle time | UAST · Cycle time | cycle time (ms): 7.0 -> 8.6 | Hurt |
Closest papers by meaning
- Open-World Drone Active Tracking with Goal-Centered Rewards (NeurIPS 2025 · similarity 0.86)
- Learning Adaptive and View-Invariant Vision Transformer for Real-Time UAV Tracking (ICML 2024 · similarity 0.85)
- Towards Realistic UAV Vision-Language Navigation: Platform, Benchmark, and Methodology (ICLR 2025 · similarity 0.85)
- MUST: The First Dataset and Unified Framework for Multispectral UAV Single Object Tracking (CVPR 2025 · similarity 0.84)
- JTD-UAV: MLLM-Enhanced Joint Tracking and Description Framework for Anti-UAV Systems (CVPR 2025 · similarity 0.84)
- UAV3D: A Large-scale 3D Perception Benchmark for Unmanned Aerial Vehicles (NeurIPS 2024 · 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.
