Research map / 3D vision / 3D generation
VisHall3D: Monocular Semantic Scene Completion from Reconstructing the Visible Regions to Hallucinating the Invisible Regions
Haoang Lu, Yuanqi Su, Xiaoning Zhang, Longjun Gao, Yu Xue, Le Wang
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 VisFrontierNet unsigned distance encoding to hard-lifting baseline (ablation) | VisHall3D baseline · SemanticKITTI val | IoU: 42.11 -> 46.44 | Improved |
| add VisFrontierNet unsigned distance encoding to hard-lifting baseline (ablation) | VisHall3D baseline · SemanticKITTI val | mIoU: 14.56 -> 15.11 | Improved |
| add OcclusionMAE without denoising (ablation) | VisFrontierNet · SemanticKITTI val | mIoU: 15.11 -> 15.99 | Improved |
| add 3D deformable attention feature refinement (ablation) | VisHall3D · SemanticKITTI val | mIoU: 15.99 -> 16.59 | Mixed |
| add denoising to OcclusionMAE (ablation) | VisHall3D · SemanticKITTI val | mIoU: 16.59 -> 17.06 | Improved |
| use occlusion-aware visibility division (theta_d 3.5 m) instead of out-of-view only (ablation) | VisHall3D · SemanticKITTI val | mIoU: 15.89 -> 17.06 | Improved |
| use relaxation threshold 3.5 m instead of 1.5 m (ablation) | VisHall3D · SemanticKITTI val | mIoU: 16.65 -> 17.06 | Improved |
| use relaxation threshold 4.5 m instead of 3.5 m (ablation) | VisHall3D · SemanticKITTI val | mIoU: 17.06 -> 16.95 | Hurt |
| add depth-direction noise (Rd=3) in OcclusionMAE (ablation) | VisHall3D · SemanticKITTI val | mIoU: 16.32 -> 16.92 | Improved |
| add horizontal noise (Rh=1) on top of depth noise Rd=3 (ablation) | VisHall3D · SemanticKITTI val | mIoU: 16.92 -> 16.52 | Mixed |
| increase depth noise from 3 to 4 voxels (ablation) | VisHall3D · SemanticKITTI val | mIoU: 16.92 -> 16.66 | Hurt |
| use two-stage VisHall3D instead of CGFormer | VisHall3D · SSCBench-KITTI-360 | mIoU: 20.05 -> 20.95 | Improved |
Later papers that compared with this method
1 later accepted paper reports VisHall3D as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Sparsity-Aware Voxel Attention and Foreground Modulation for 3D Semantic Scene Completion (CVPR 2026 · better on 2 of 2 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- MonoScene — MonoScene: Monocular 3D Semantic Scene Completion (CVPR 2022) · better on 5 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- MonoScene: Monocular 3D Semantic Scene Completion (CVPR 2022 · similarity 0.90)
- See and Think: Disentangling Semantic Scene Completion (NeurIPS 2018 · similarity 0.90)
- Unleashing Semantic and Geometric Priors for 3D Scene Completion (AAAI 2026 · similarity 0.89)
- VLScene: Vision-Language Guidance Distillation for Camera-Based 3D Semantic Scene Completion (AAAI 2025 · similarity 0.89)
- Bi-SSC: Geometric-Semantic Bidirectional Fusion for Camera-based 3D Semantic Scene Completion (CVPR 2024 · similarity 0.88)
- Memory-Augmented Re-Completion for 3D Semantic Scene Completion (AAAI 2025 · 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.
