Research map / Point clouds / Segmentation and detection
Not All Voxels Are Equal: Semantic Scene Completion from the Point-Voxel Perspective
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 |
|---|---|---|---|
| add AVA to the point-based baseline (ablation) | PointNet++ baseline · NYUCAD test | SSC mIoU: 51.4 -> 54.9 | Helped |
| add SP to the point-based baseline (ablation) | PointNet++ baseline · NYUCAD test | SSC mIoU: 51.4 -> 53.2 | Helped |
| add both AVA and SP (ablation) | PointNet++ baseline · NYUCAD test | SSC mIoU: 51.4 -> 56.9 | Helped |
| replace point-only processing with nearest-voxel aggregation (ablation) | PVA-Net · NYUCAD test | SSC mIoU: 53.2 -> 55.1 | Helped |
| replace point-only processing with spherical voxel aggregation (ablation) | PVA-Net · NYUCAD test | SSC mIoU: 53.2 -> 56.0 | Helped |
| use anisotropic rather than spherical receptive fields (ablation) | PVA-Net · NYUCAD test | SSC mIoU: 56.0 -> 56.9 | Helped |
| move AVA from SA-1 to SA-2 (ablation) | PVA-Net · NYUCAD test | SSC mIoU: 56.9 -> 55.7 | Hurt |
| move AVA from SA-1 to SA-3 (ablation) | PVA-Net · NYUCAD test | SSC mIoU: 56.9 -> 53.4 | Hurt |
| move AVA from SA-1 to SA-4 (ablation) | PVA-Net · NYUCAD test | SSC mIoU: 56.9 -> 53.8 | Hurt |
| apply AVA at every SA layer instead of only SA-1 (ablation) | PVA-Net · NYUCAD test | SSC mIoU: 56.9 -> 56.6 | Hurt |
| replace semantic-guided propagation with inverse Euclidean weights (ablation) | PVA-Net · NYUCAD test | SSC mIoU: 56.9 -> 54.9 | Hurt |
| replace semantic-guided propagation with cosine-similarity weights (ablation) | PVA-Net · NYUCAD test | SSC mIoU: 56.9 -> 54.4 | Hurt |
Closest papers by meaning
- Voxel Proposal Network via Multi-Frame Knowledge Distillation for Semantic Scene Completion (NeurIPS 2024 · similarity 0.92)
- Context and Geometry Aware Voxel Transformer for Semantic Scene Completion (NeurIPS 2024 · similarity 0.91)
- H2GFormer: Horizontal-to-Global Voxel Transformer for 3D Semantic Scene Completion (AAAI 2024 · similarity 0.90)
- Not All Voxels Are Equal: Hardness-Aware Semantic Scene Completion with Self-Distillation (CVPR 2024 · similarity 0.90)
- Memory-Augmented Re-Completion for 3D Semantic Scene Completion (AAAI 2025 · similarity 0.89)
- Learning Temporal 3D Semantic Scene Completion via Optical Flow Guidance (NeurIPS 2025 · similarity 0.89)
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.