Research map / 3D vision / Gaussian splatting
Submodular Trajectory Optimization for Aerial 3D Scanning
Mike Roberts, Debadeepta Dey, Anh Truong, Sudipta Sinha, Shital Shah, Ashish Kapoor, Pat Hanrahan, Neel Joshi
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 |
|---|---|---|---|
| plan with submodular coverage orienteering instead of greedy next-best-view | aerial trajectory planner · Grass Lands synthetic scene (960 m budget) | completeness error (mm): 330.7 -> 323.3 | Improved |
| plan with submodular coverage orienteering instead of greedy next-best-view | aerial trajectory planner · Grass Lands synthetic scene (960 m budget) | accuracy error (mm): 122.8 -> 115.2 | Improved |
| plan with submodular coverage orienteering instead of greedy next-best-view | aerial trajectory planner · Grass Lands synthetic scene (960 m budget) | visual error (%): 3.6 -> 3.3 | Improved |
| plan with scene-aware coverage trajectory instead of overhead orbit plus lawnmower | aerial trajectory planner · Grass Lands synthetic scene (960 m budget) | completeness error (mm): 583.8 -> 323.3 | Improved |
| plan with scene-aware coverage trajectory instead of overhead orbit plus lawnmower | aerial trajectory planner · Grass Lands synthetic scene (960 m budget) | accuracy error (mm): 170.2 -> 115.2 | Improved |
| plan with scene-aware coverage trajectory instead of overhead orbit plus lawnmower | aerial trajectory planner · Grass Lands synthetic scene (960 m budget) | visual error (%): 7.1 -> 3.3 | Improved |
| optimize coverage-based trajectory instead of random graph-node visits | aerial trajectory planner · Grass Lands synthetic scene (960 m budget) | completeness error (mm): 557.2 -> 323.3 | Improved |
| optimize coverage-based trajectory instead of random graph-node visits | aerial trajectory planner · Grass Lands synthetic scene (960 m budget) | accuracy error (mm): 126.5 -> 115.2 | Improved |
| use our additive-approximation ILP orienteering instead of next-best-view greedy | submodular orienteering · Synthetic scene, 960 m budget | submodular reward: 20% more reward | Improved |
| use our ILP orienteering instead of next-best-view greedy | submodular orienteering · Synthetic scene | computation time | Hurt |
| use p-SPIEL orienteering instead of our method | submodular orienteering · Synthetic scene | solution success and time: over 4 hours | Hurt |
Closest papers by meaning
- AirZoo: A Unified Large-Scale Dataset for Grounding Aerial Geometric 3D Vision (ECCV 2026 · similarity 0.87)
- DroneSplat: 3D Gaussian Splatting for Robust 3D Reconstruction from In-the-Wild Drone Imagery (CVPR 2025 · similarity 0.86)
- AeroGS: Scale-Aware Gaussian Splatting for Pose-Free Dynamic UAV Scene Reconstruction (CVPR 2026 · similarity 0.85)
- DroneSplat+: Semantics-Enhanced 3D Gaussian Splatting for Robust 3D Reconstruction From In-the-Wild Drone Imagery (TPAMI 2026 · similarity 0.85)
- UAV3D: A Large-scale 3D Perception Benchmark for Unmanned Aerial Vehicles (NeurIPS 2024 · similarity 0.83)
- Seeing Across Skies and Streets: Feedforward 3D Reconstruction from Satellite, Drone, and Ground Images (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.
