Research map / Vision-language and multimodal / Robot action and navigation
OpenFly: A COMPREHENSIVE PLATFORM FOR AERIAL VISION-LANGUAGE NAVIGATION
Yunpeng Gao, Chenhui Li, Zhongrui You, Junli Liu, Li Zhen, Pengan CHEN, Qizhi Chen, Zhonghan Tang, Liansheng Wang, Yangpenghui, Yiwen Tang, Yuhang Tang and 11 more
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 historical frames (uniform sampling) to OpenVLA (ablation) | OpenVLA · OpenFly test-seen | SR (%): 2.3 -> 6.9 | Improved |
| add keyframe selection with token merging versus uniform history with token merging (ablation) | OpenFly-Agent · OpenFly test-seen | SR (%): 16.6 -> 34.3 | Improved |
| add visual token merging to keyframe selection (ablation) | OpenFly-Agent · OpenFly test-seen | SR (%): 9.2 -> 34.3 | Improved |
| add visual token merging to keyframe selection (ablation) | OpenFly-Agent · OpenFly test-seen | NE (m): 275 -> 93 | Improved |
| select keyframes by landmark rather than at random (ablation) | OpenFly-Agent · OpenFly test-seen | SR (%): 8.7 -> 9.2 | Improved |
| keyframe selection alone raises navigation error (ablation) | OpenFly-Agent · OpenFly test-seen | NE (m): 223 -> 275 | Hurt |
| OpenFly-Agent vs NaVila | OpenFly-Agent · OpenFly test-seen | SR (%): 20.3 -> 34.3 | Improved |
| OpenFly-Agent vs NaVila | OpenFly-Agent · OpenFly test-unseen | SR (%): 14.7 -> 22.6 | Improved |
| generalize from seen to unseen scenes | OpenFly-Agent · OpenFly | SR (%): 34.3 -> 22.6 | Hurt |
| train on OpenFly instead of AerialVLN | OpenFly-Agent · real-world flights | SR (%): 4.3 -> 26.09 | Improved |
Closest papers by meaning
- Towards Realistic UAV Vision-Language Navigation: Platform, Benchmark, and Methodology (ICLR 2025 · similarity 0.93)
- SatNav: A Scalable Benchmark for Long-Horizon UAV Vision-Language Navigation from Satellite Imagery (NeurIPS 2026 · similarity 0.90)
- WorldVLN: Autoregressive World Action Model for Aerial Vision-Language Navigation (NeurIPS 2026 · similarity 0.90)
- Is your VLM Sky-Ready? A Comprehensive Spatial Intelligence Benchmark for UAV Navigation (CVPR 2026 · similarity 0.89)
- FlySearch: Exploring how vision-language models explore (NeurIPS 2025 · similarity 0.89)
- Learning Fine-Grained Alignment for Aerial Vision-Dialog Navigation (AAAI 2025 · similarity 0.86)
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.
