Research map / 3D vision / 3D generation
AirSim360: A Panoramic Simulation Platform within Drone View
Xian Ge, Yuling Pan, Yuhang Zhang, Xiang Li, Weijun Zhang, Dizhe Zhang, Zhaoliang Wan, Xin Lin, Xiangkai Zhang, Juntao Liang, Xiangtai Li, WenJie Jiang and 3 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 |
|---|---|---|---|
| disable Capture Every Frame on rendering cameras (ablation) | AirSim360 · frame capture | FPS: 20 -> 29 | Improved |
| disable Capture Every Frame on rendering cameras (ablation) | AirSim360 · frame capture | GPU time: 54 ms -> 35 ms | Improved |
| use RHI GPU-side texture copy stitching of six cube faces (ablation) | AirSim360 · 6-camera panoramic capture | FPS: 14 -> 18 | Improved |
| add Omni360-Human synthetic data to nuScenes training | MPDE model · MPDE public test sets | angular error: 21.21 -> 17.02 | Improved |
| add Omni360-Human synthetic data to nuScenes training | MPDE model · MPDE public test sets | distance error: 0.484 -> 0.458 | Improved |
| add Omni360-Human synthetic data | MPDE model · FreeMan | angular error: 17 -> 11.6 | Improved |
| add Omni360-Human synthetic data | MPDE model · KITTI | angular error: 31.50 -> 31.20 | No clear effect |
| use pitch 20 subset instead of pitch 0 subset of Omni360-Human (ablation) | MPDE model · MPDE public test sets | angular error: 19.194 -> 17.023 | Improved |
| train UniK3D on Omni360 instead of Deep360 | UniK3D · SphereCraft out-of-domain | AbsRel: 8.2570 -> 5.4372 | Improved |
| train UniK3D on Omni360 instead of Deep360 | UniK3D · cross-domain | AbsRel: 0.3600 -> 0.1762 | Improved |
| add Omni360-Scene data to WildPASS training | OOOPS semantic segmentation · WildPASS val | mIoU: 58.0 -> 67.4 | Improved |
| add Omni360-Scene data to WildPASS training | Mask2Former entity segmentation · WildPASS val | mAP: 24.6 -> 38.9 | Improved |
Closest papers by meaning
- 360+x: A Panoptic Multi-modal Scene Understanding Dataset (CVPR 2024 · similarity 0.85)
- ODI-Bench: Can MLLMs Understand Immersive Omnidirectional Environments? (ICLR 2026 · similarity 0.85)
- AirCopBench: A Benchmark for Multi-drone Collaborative Embodied Perception and Reasoning (AAAI 2026 · similarity 0.85)
- Image2Sim: Scaling Embodied Navigation via Generative Neural Simulator (NeurIPS 2026 · similarity 0.84)
- AGC-Drive: A Large-Scale Dataset for Real-World Aerial-Ground Collaboration in Driving Scenarios (NeurIPS 2025 · similarity 0.84)
- Is your VLM Sky-Ready? A Comprehensive Spatial Intelligence Benchmark for UAV Navigation (CVPR 2026 · 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.
