Research map / 3D vision / 3D generation
Learning View Selection for 3D Scenes
Yifan Sun, Qixing Huang, Dun-Yu Hsiao, Li Guan, Gang Hua
Check 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 |
|---|---|---|---|
| Use learned view prediction and continuous camera-pose optimization instead of dense-grid selection | 3D scene view selection · 339-scene test benchmark, 10 cameras | S-Avg.: 0.711 -> 0.724 | Improved |
| Use learned view prediction and continuous camera-pose optimization instead of dense-grid selection | 3D scene view selection · 339-scene test benchmark, 80% coverage | number of cameras: 22.3 -> 18.8 | Improved |
| Use learned view prediction and continuous camera-pose optimization instead of dense-grid selection | 3D scene view selection · 339-scene test benchmark, 90% coverage | number of cameras: 40.4 -> 34.2 | Improved |
| Use learned view prediction and continuous camera-pose optimization instead of dense-grid selection | 3D scene view selection · 339-scene test benchmark, matched performance | time (sec.): 4873 -> 462 | Improved |
| Increase the number of initial camera-grid samples (ablation) | Ours · 3D scene view selection | coverage performance | No clear effect |
| Vary the scoring-function hyperparameters (ablation) | Ours · 339-scene test benchmark, 80% coverage | number of cameras: ±0.6 #cameras | Mixed |
| Use different camera-pose optimization strategies instead of BFGS (ablation) | Ours · 3D scene view selection | optimal solution | No clear effect |
| Use different scene-encoding resolutions (ablation) | Ours · 3D scene view selection | performance | No clear effect |
Closest papers by meaning
- Coverage Optimization for Camera View Selection (CVPR 2026 · similarity 0.88)
- Learning 3D Object Shape and Layout Without 3D Supervision (CVPR 2022 · similarity 0.86)
- Learning 3D Scene Priors With 2D Supervision (CVPR 2023 · similarity 0.86)
- SCONE: Surface Coverage Optimization in Unknown Environments by Volumetric Integration (NeurIPS 2022 · similarity 0.86)
- Improving Viewpoint-Independent Object-Centric Representations through Active Viewpoint Selection (NeurIPS 2024 · similarity 0.85)
- OSN: Infinite Representations of Dynamic 3D Scenes from Monocular Videos (ICML 2024 · similarity 0.85)
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.
