Research map / Detection and segmentation / 3D detection and driving
RevealNet: Seeing Behind Objects in RGB-D Scans
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| jointly predict complete object geometry end-to-end rather than chaining separate completion and segmentation systems | RevealNet · ScanNet semantic instance completion | mAP@0.5: 5.49 -> 21.77 | Helped |
| jointly predict complete object geometry end-to-end rather than chaining separate completion and segmentation systems | RevealNet · SUNCG semantic instance completion | mAP@0.5: 5.5 -> 24.0 | Helped |
| add instance completion to the detection and segmentation model (ablation) | RevealNet · ScanNet 3D detection | mAP@0.5: 31.93 -> 36.39 | Helped |
| add instance completion to the detection and segmentation model (ablation) | RevealNet · ScanNet instance segmentation | mAP@0.5: 24.49 -> 30.52 | Helped |
| add instance completion to the detection and segmentation model (ablation) | RevealNet · SUNCG 3D detection | mAP@0.5: 29.80 -> 37.81 | Helped |
| add instance completion to the detection and segmentation model (ablation) | RevealNet · SUNCG instance segmentation | mAP@0.5: 23.86 -> 36.28 | Helped |
| add RGB color features (ablation) | RevealNet · ScanNet semantic instance completion | mAP@0.5: 14.52 -> 21.77 | Helped |
| add RGB color features (ablation) | RevealNet · SUNCG semantic instance completion | mAP@0.5: 20.5 -> 24.0 | Helped |
| add the global geometric completion proxy loss (ablation) | RevealNet · ScanNet semantic instance completion | mAP@0.5: 17.03 -> 21.77 | Helped |
| add the global geometric completion proxy loss (ablation) | RevealNet · SUNCG semantic instance completion | mAP@0.5: 20.9 -> 24.0 | Helped |
| add RGB color features (ablation) | RevealNet · SUNCG 3D detection | mAP@0.5: 31.75 -> 37.81 | Hurt |
| add RGB color features (ablation) | RevealNet · SUNCG instance segmentation | mAP@0.5: 31.59 -> 36.28 | Helped |
Closest papers by meaning
- Towards Part-Based Understanding of RGB-D Scans (CVPR 2021 · similarity 0.91)
- Semantic Scene Completion via Integrating Instances and Scene In-the-Loop (CVPR 2021 · similarity 0.91)
- Open3DIS: Open-Vocabulary 3D Instance Segmentation with 2D Mask Guidance (CVPR 2024 · similarity 0.89)
- ImageNet3D: Towards General-Purpose Object-Level 3D Understanding (NeurIPS 2024 · similarity 0.89)
- Point-based Instance Completion with Scene Constraints (ICLR 2025 · similarity 0.88)
- RefineAny3D: Depth Refinement as Semantic Alignment for Monocular 3D Detection (NeurIPS 2026 · similarity 0.88)
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.