Research map / 3D vision / 3D generation
AirZoo: A Unified Large-Scale Dataset for Grounding Aerial Geometric 3D Vision
Xiaoya Cheng, Rouwan Wu, Xinyi Liu, Zeyu Cui, Yan Liu, Na Zhao, Yu Liu, Maojun Zhang, Shen Yan
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 |
|---|---|---|---|
| fine-tune on AirZoo | MegaLoc · UAV-VisLoc (real, zero-shot) | Recall@1: 26.91 -> 28.64 | Improved |
| fine-tune on AirZoo | MegaLoc · AirZoo-Real geo-localization (real, zero-shot) | Recall@1: 6.46 -> 18.66 | Improved |
| fine-tune on AirZoo | MegaLoc · AirZoo-Real geo-localization (real, zero-shot) | Recall@5: 22.16 -> 50.53 | Improved |
| fine-tune on AirZoo | MegaLoc · UAV-VisLoc Scene 7 | Recall@1: 38.46 -> 26.92 | Hurt |
| fine-tune on AirZoo (with MegaDepth) | RoMa · AerialExtreMatch 6-DoF localization (5m, 1deg) | pose recall (5m, 1deg): 95.83 -> 96.21 | Improved |
| fine-tune on AirZoo (with MegaDepth) | RoMa · AirZoo-Real translation matching | acc@5m: 44.95 -> 79.78 | Improved |
| fine-tune on AirZoo (with MegaDepth) | RoMa · AirZoo-Real translation matching | median translation error (m): 5.87 -> 3.03 | Improved |
| use AirZoo fine-tuning instead of GIM video-trained RoMa | RoMa · AerialExtreMatch 6-DoF localization (5m, 1deg) | pose recall (5m, 1deg): 94.32 -> 96.21 | Improved |
| fine-tune on AirZoo | VGGT · UrbanScene3D | F1: 43.38 -> 52.61 | Improved |
| fine-tune on AirZoo | VGGT · AirZoo-Test (synthetic) | F1: 67.67 -> 77.32 | Improved |
| fine-tune on AirZoo | VGGT · UAVScenes | F1: 86.20 -> 86.98 | Improved |
| fine-tune on AirZoo | VGGT · AirZoo-Real reconstruction | F1: 42.48 -> 43.67 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- VGGT — VGGT: Visual Geometry Grounded Transformer (CVPR 2025) · better on 4 of 4 shared results
- DUSt3R — DUSt3R: Geometric 3D Vision Made Easy (CVPR 2024) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- UAV3D: A Large-scale 3D Perception Benchmark for Unmanned Aerial Vehicles (NeurIPS 2024 · similarity 0.89)
- AerialMegaDepth: Learning Aerial-Ground Reconstruction and View Synthesis (CVPR 2025 · similarity 0.88)
- Seeing Across Skies and Streets: Feedforward 3D Reconstruction from Satellite, Drone, and Ground Images (NeurIPS 2026 · similarity 0.88)
- WildBox: A Dataset and Benchmark for Aerial Monocular 3D Detection of African Savanna Wildlife (NeurIPS 2026 · similarity 0.87)
- AGC-Drive: A Large-Scale Dataset for Real-World Aerial-Ground Collaboration in Driving Scenarios (NeurIPS 2025 · similarity 0.87)
- USCILab3D: A Large-scale, Long-term, Semantically Annotated Outdoor Dataset (NeurIPS 2024 · similarity 0.87)
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.
