Research map / Video / Action recognition
UAV-Human: A Large Benchmark for Human Behavior Understanding With Unmanned Aerial Vehicles
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 guided transformers trained with flat RGB feature guidance to fisheye I3D | Fisheye action recognition on UAV-Human · UAV-Human CSv1 | accuracy (%): 20.76 -> 23.24 | Helped |
| Add RGB guidance loss to the fisheye stream without video transformers (ablation) | Fisheye I3D · UAV-Human CSv1 | accuracy (%): 20.76 -> 21.68 | Helped |
| Add video transformers without RGB guidance (ablation) | Fisheye I3D · UAV-Human CSv1 | accuracy (%): 20.76 -> 21.49 | Helped |
| Combine RGB feature guidance and video transformers (ablation) | Fisheye I3D · UAV-Human CSv1 | accuracy (%): 21.68 -> 23.24; 21.49 -> 23.24 | Helped |
| Use skeleton representations instead of video inputs | Action recognition on UAV-Human · UAV-Human action recognition | accuracy | Helped |
| Use night-vision or IR rather than other tested modalities | I3D action recognition · UAV-Human CSv1 | accuracy (%): 28.72; 26.56 | Helped |
Closest papers by meaning
- 3D-aware Select, Expand, and Squeeze Token for Aerial Action Recognition (AAAI 2025 · similarity 0.86)
- UAV3D: A Large-scale 3D Perception Benchmark for Unmanned Aerial Vehicles (NeurIPS 2024 · similarity 0.86)
- AG-VPReID: A Challenging Large-Scale Benchmark for Aerial-Ground Video-based Person Re-Identification (CVPR 2025 · similarity 0.85)
- Towards Realistic UAV Vision-Language Navigation: Platform, Benchmark, and Methodology (ICLR 2025 · similarity 0.84)
- Multiview Aerial Visual RECognition (MAVREC): Can Multi-view Improve Aerial Visual Perception? (CVPR 2024 · similarity 0.83)
- M$^2$E-UAV: A Benchmark and Analysis for Onboard Motion-on-Motion Event-Based Tiny UAV Detection (NeurIPS 2026 · similarity 0.83)
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.