Research map / 3D vision / 3D generation
MM-3DScene: 3D Scene Understanding by Customizing Masked Modeling With Informative-Preserved Reconstruction and Self-Distilled Consistency
Read the paperCheck 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 |
|---|---|---|---|
| pre-train with MM-3DScene instead of training from scratch | VoteNet · ScanNetv2 | mAP@0.5: 35.4 -> 41.5 | Helped |
| pre-train with MM-3DScene instead of training from scratch | VoteNet · SUN RGB-D | mAP@0.5: 32.9 -> 37.3 | Helped |
| pre-train with MM-3DScene instead of training from scratch | Point Transformer · ScanNetv2 | mIoU: 70.6 -> 72.8 | Helped |
| pre-train with MM-3DScene instead of training from scratch | Point Transformer · S3DIS Area-5 | mIoU: 70.4 -> 71.9 | Helped |
| add informative-preserved masked modeling to consistency self-distillation (ablation) | MM-3DScene · S3DIS Area-5 | mIoU: 70.9 -> 71.9 | Helped |
| add consistency self-distillation to informative-preserved masked modeling (ablation) | MM-3DScene · S3DIS Area-5 | mIoU: 71.1 -> 71.9 | Helped |
| add progressive reconstruction to random masking (ablation) | PointMAE* · S3DIS Area-5 | mIoU: 70.4 -> 70.6 | Helped |
| preserve informative points rather than mask them (ablation) | MM-3DScene masked modeling · S3DIS Area-5 | mIoU: 70.2 -> 70.9 | Helped |
| add progressive reconstruction to informative-preserved masking (ablation) | MM-3DScene (MM only) · S3DIS Area-5 | mIoU: 70.9 -> 71.1 | Helped |
| use coordinate and RGB local differences rather than coordinate differences only (ablation) | MM-3DScene (MM only) · S3DIS Area-5 | mIoU: 70.6 -> 71.1 | Helped |
| use coordinate and RGB local differences rather than RGB differences only (ablation) | MM-3DScene (MM only) · S3DIS Area-5 | mIoU: 70.9 -> 71.1 | Helped |
| increase the reconstruction gap beyond 0.1 (ablation) | MM-3DScene (MM only) · S3DIS | mIoU | Hurt |
Closest papers by meaning
- Masked Scene Modeling: Narrowing the Gap Between Supervised and Self-Supervised Learning in 3D Scene Understanding (CVPR 2025 · similarity 0.88)
- Any3DIS: Class-Agnostic 3D Instance Segmentation by 2D Mask Tracking (CVPR 2025 · similarity 0.87)
- MaskGaussian: Adaptive 3D Gaussian Representation from Probabilistic Masks (CVPR 2025 · similarity 0.86)
- Open3DIS: Open-Vocabulary 3D Instance Segmentation with 2D Mask Guidance (CVPR 2024 · similarity 0.86)
- Bootstraping Clustering of Gaussians for View-consistent 3D Scene Understanding (AAAI 2025 · similarity 0.86)
- Geometry-Guided Semantic Reconstruction for 3D Instance Segmentation (NeurIPS 2026 · similarity 0.86)
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.