Research map / 3D vision / 3D generation
Paparazzo: Active Mapping of Moving 3D Objects
Davide Allegro, Shiyao Li, Stefano Ghidoni, Vincent Lepetit
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 synchronization-aware viewpoint selection (cost C_sync) to random informative selection (ablation) | Random Informative Selection · Active mapping, Bouncing Ball motion (6 Habitat scenes, avg) | 3D coverage (%): 67.07 -> 81.51 | Improved |
| add synchronization-aware viewpoint selection (cost C_sync) to random informative selection (ablation) | Random Informative Selection · Active mapping, Stop & Go motion (6 Habitat scenes, avg) | 3D coverage (%): 48.58 -> 68.03 | Improved |
| add active mapping mode with EKF-predicted informative viewpoints on top of passive tracking (ablation) | Tracking-Only · Active mapping, Bouncing Ball motion (6 Habitat scenes, avg) | 3D coverage (%): 75.89 -> 81.51 | Improved |
| add active mapping mode on top of passive tracking (ablation) | Tracking-Only · Active mapping, Forward & Backward motion (6 Habitat scenes, avg) | 3D coverage (%): 62.05 -> 67.73 | Improved |
| add active mapping mode on top of passive tracking (ablation) | Tracking-Only · Active mapping, Stop & Go motion (6 Habitat scenes, avg) | 3D coverage (%): 60.79 -> 68.03 | Improved |
| add active mapping mode on top of passive tracking (ablation) | Tracking-Only · Active mapping, Stop & Go motion (6 Habitat scenes, avg) | AUC of coverage vs steps: 0.59 -> 0.65 | Improved |
| add active mapping mode on top of passive tracking (ablation) | Tracking-Only · Active mapping, Curved Bouncing Ball motion (6 Habitat scenes, avg) | 3D coverage (%): 68.15 -> 70.73 | Improved |
| add active mapping mode on top of passive tracking (ablation) | Tracking-Only · Active mapping, Curved Bouncing Ball motion (6 Habitat scenes, avg) | completeness (cm): 1.11 -> 1.11 | No clear effect |
| use Paparazzo instead of Tracking-Only in the narrow Ribera scene with curved motion (ablation) | Tracking-Only · Ribera scene, Curved Bouncing Ball | 3D coverage (%): 70.84 -> 67.52 | Hurt |
| random informative viewpoint selection vs random walk under Stop & Go motion | Random Walk · Active mapping, Stop & Go motion (6 Habitat scenes, avg) | 3D coverage (%): 51.80 -> 48.58 | Hurt |
| random informative viewpoint selection vs random walk under Bouncing Ball motion | Random Walk · Active mapping, Bouncing Ball motion (6 Habitat scenes, avg) | 3D coverage (%): 51.50 -> 67.07 | Improved |
| Paparazzo vs Tracking-Only on Stop & Go with Object 1 | Tracking-Only · Stop & Go, Object 1 | 3D coverage (%): 68.25 -> 81.23 | Improved |
Closest papers by meaning
- NextBestPath: Efficient 3D Mapping of Unseen Environments (ICLR 2025 · similarity 0.86)
- ActiveGAMER: Active GAussian Mapping through Efficient Rendering (CVPR 2025 · similarity 0.85)
- GS-SLAM: Dense Visual SLAM with 3D Gaussian Splatting (CVPR 2024 · similarity 0.85)
- Gaussian Splatting SLAM (CVPR 2024 · similarity 0.85)
- Stereo4D: Learning How Things Move in 3D from Internet Stereo Videos (CVPR 2025 · similarity 0.84)
- Understanding while Exploring: Semantics-driven Active Mapping (NeurIPS 2025 · 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.
