Research map / 3D vision / 3D generation
X-Section: Cross-Section Prediction for Enhanced RGB-D Fusion
Andrea Nicastro, Ronald Clark, Stefan Leutenegger
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 depth+silhouette input instead of RGB-D input (ablation) | X-Section · YCB-Video thickness prediction | absolute relative difference (ResNet101): 4.301 -> 3.819 | Improved |
| use ResNet101 instead of ResNet50 backbone (ablation) | X-Section · YCB-Video thickness prediction | RMSE (linear), DS input: 0.013 vs 0.015 | No clear effect |
| add depth+silhouette input instead of RGB-D input (ablation) | X-Section · Voxlets dataset (all objects detected) | IoU (ResNet101): 0.620 -> 0.761 | Improved |
| add thickness prediction to TSDF fusion | X-Section · Voxlets dataset (all objects detected) | IoU: 0.327 -> 0.761 | Improved |
| add thickness-enhanced TSDF fusion over depth-only fusion | X-Section · YCB Video sequence 0051 | IoU: 0.355 -> 0.735 | Improved |
| add thickness-enhanced TSDF fusion over depth-only fusion | X-Section · YCB Video sequence 0052 | IoU: 0.264 -> 0.693 | Improved |
| add thickness-enhanced TSDF fusion over depth-only fusion | X-Section · YCB Video sequence 0049 | precision: 0.745 -> 0.659 | Hurt |
| add thickness-enhanced TSDF fusion over depth-only fusion | X-Section · Voxlets dataset (missed detections) | IoU: 0.234 -> 0.440 | Improved |
| use X-Section instead of Voxlets when detector misses objects | X-Section · Voxlets dataset (missed detections) | IoU: 0.622 vs 0.440 | Hurt |
Closest papers by meaning
- Object-X: Learning to Reconstruct Multi-Modal 3D Object Representations (NeurIPS 2025 · similarity 0.85)
- CoReNet: Coherent 3D Scene Reconstruction from a Single RGB Image (ECCV 2020 · similarity 0.85)
- OReX: Object Reconstruction From Planar Cross-Sections Using Neural Fields (CVPR 2023 · similarity 0.85)
- DI-Fusion: Online Implicit 3D Reconstruction With Deep Priors (CVPR 2021 · similarity 0.84)
- UniSDF: Unifying Neural Representations for High-Fidelity 3D Reconstruction of Complex Scenes with Reflections (NeurIPS 2024 · similarity 0.84)
- RaySt3R: Predicting Novel Depth Maps for Zero-Shot Object Completion (NeurIPS 2025 · 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.
