Research map / 3D vision / 3D generation
MonoScene: Monocular 3D Semantic Scene Completion
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 |
|---|---|---|---|
| Use single-image RGB MonoScene instead of the strongest RGB-inferred baseline | NYUv2 test · NYUv2 test | mIoU: 22.91 -> 26.94 | Helped |
| Use single-image RGB MonoScene instead of the strongest RGB-inferred baseline | SemanticKITTI hidden test · SemanticKITTI hidden test | mIoU: 8.97 -> 11.08 | Helped |
| Add FLoSP in place of interpolated and convolved 2D features (ablation) | MonoScene · NYUv2 test | mIoU: 14.11 -> 26.94 | Helped |
| Add FLoSP in place of interpolated and convolved 2D features (ablation) | MonoScene · SemanticKITTI validation | IoU: 27.55 -> 37.12 | Helped |
| Add 3D CRP (ablation) | MonoScene · NYUv2 test | mIoU: 26.27 -> 26.94 | Helped |
| Add semantic Scene-Class Affinity Loss (ablation) | MonoScene · NYUv2 test | IoU: 42.82 -> 42.51 | Hurt |
| Add semantic Scene-Class Affinity Loss (ablation) | MonoScene · SemanticKITTI validation | IoU: +0.34 | Helped |
| Add geometry Scene-Class Affinity Loss (ablation) | MonoScene · NYUv2 test | IoU: +1.55 | Helped |
| Add geometry Scene-Class Affinity Loss (ablation) | MonoScene · SemanticKITTI validation | IoU: +2.20 | Helped |
| Add Frustum Proportion Loss (ablation) | MonoScene · NYUv2 test | mIoU: 26.37 -> 26.94 | Helped |
| Use more FLoSP scales, adding scales 4 and 8 to 1 and 2 (ablation) | MonoScene · NYUv2 test | mIoU: 25.66 -> 26.94 | Helped |
| Use four relations instead of two in 3D CRP (ablation) | MonoScene · NYUv2 test | mIoU: 26.63 -> 26.94 | Helped |
Closest papers by meaning
- Bi-SSC: Geometric-Semantic Bidirectional Fusion for Camera-based 3D Semantic Scene Completion (CVPR 2024 · similarity 0.91)
- Know Your Neighbors: Improving Single-View Reconstruction via Spatial Vision-Language Reasoning (CVPR 2024 · similarity 0.88)
- IPFormer: Visual 3D Panoptic Scene Completion with Context-Adaptive Instance Proposals (NeurIPS 2025 · similarity 0.88)
- H2GFormer: Horizontal-to-Global Voxel Transformer for 3D Semantic Scene Completion (AAAI 2024 · similarity 0.88)
- Tackling View-Dependent Semantics in 3D Language Gaussian Splatting (ICML 2025 · similarity 0.88)
- Semi-supervised 3D Semantic Scene Completion with 2D Vision Foundation Model Guidance (AAAI 2025 · similarity 0.87)
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.