Research map / Point clouds / Segmentation and detection
Weakly Supervised Semantic Segmentation for Large-Scale Point Cloud
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 self-supervised pre-training rather than training from scratch (ablation) | RandLA-Net weakly supervised segmentation · S3DIS Area5, 1pt | mIoU: 40.4 -> 45.4 | Helped |
| Add self-supervised pre-training rather than training from scratch (ablation) | RandLA-Net weakly supervised segmentation · S3DIS Area5, SPT | mIoU: 55.8 -> 57.5 | Helped |
| Add self-supervised pre-training rather than training from scratch (ablation) | RandLA-Net weakly supervised segmentation · S3DIS Area5, 1% | mIoU: 58.6 -> 60.1 | Helped |
| Add self-supervised pre-training rather than training from scratch (ablation) | RandLA-Net weakly supervised segmentation · S3DIS Area5, 10% | mIoU: 60.6 -> 61.3 | Helped |
| Add nonlinear-weighted sparse label propagation to pre-training with λ = 0 (ablation) | RandLA-Net weakly supervised segmentation · S3DIS Area5, 1pt | mIoU: 45.4 -> 45.8 | Helped |
| Add nonlinear-weighted sparse label propagation to pre-training with λ = 0 (ablation) | RandLA-Net weakly supervised segmentation · S3DIS Area5, SPT | mIoU: 57.5 -> 58.2 | Helped |
| Add nonlinear-weighted sparse label propagation to pre-training with λ = 0 (ablation) | RandLA-Net weakly supervised segmentation · S3DIS Area5, 1% | mIoU: 60.1 -> 61.8 | Helped |
| Add nonlinear-weighted sparse label propagation to pre-training with λ = 0 (ablation) | RandLA-Net weakly supervised segmentation · S3DIS Area5, 10% | mIoU: 61.3 -> 64.0 | Helped |
| Use a constant λ = 1 instead of ramping the propagation loss nonlinearly (ablation) | RandLA-Net weakly supervised segmentation · S3DIS Area5 | mIoU | Hurt |
Closest papers by meaning
- Less Is More: Label Recommendation for Weakly Supervised Point Cloud Semantic Segmentation (AAAI 2024 · similarity 0.91)
- Weakly Supervised Point Cloud Semantic Segmentation via Artificial Oracle (CVPR 2024 · similarity 0.90)
- Point Cloud Semantic Segmentation with Sparse and Inhomogeneous Annotations (AAAI 2025 · similarity 0.90)
- Multi-Modality Affinity Inference for Weakly Supervised 3D Semantic Segmentation (AAAI 2024 · similarity 0.89)
- Distribution Guidance Network for Weakly Supervised Point Cloud Semantic Segmentation (NeurIPS 2024 · similarity 0.88)
- Generative Hard Example Augmentation for Semantic Point Cloud Segmentation (CVPR 2025 · 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.