Research map / Detection and segmentation / Open-set and few-shot detection
Novel Class Discovery Meets Foundation Models for 3D Semantic Segmentation
Luigi Riz, Cristiano Saltori, Yiming Wang, Elisa Ricci, Fabio Poiesi
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 semantic alignment with a 3D foundation model (SNOPS over NOPS) | NOPS · SemanticPOSS | novel mIoU: 21.40 -> 30.05 | Improved |
| add semantic alignment and online pseudo-labelling (SNOPS over EUMS) | EUMS+ · SemanticPOSS | novel mIoU: 14.94 -> 30.05 | Improved |
| add semantic alignment with foundation model | NOPS · SemanticKITTI | novel mIoU: 22.84 -> 26.39 | Improved |
| add semantic alignment with foundation model | NOPS · S3DIS | novel mIoU: 20.79 -> 34.05 | Improved |
| add semantic alignment with foundation model | NOPS · S3DIS-4^0 | novel mIoU: 26.08 -> 55.94 | Improved |
| add semantic alignment with foundation model | NOPS · S3DIS-3^3 | novel mIoU: 12.99 -> 11.24 | Hurt |
| add semantic alignment with foundation model | NOPS · SemanticPOSS bike class | IoU: 35.47 -> 34.21 | Hurt |
| remove pre-training on base classes (train from scratch) (ablation) | SNOPS ablation · SemanticPOSS | novel mIoU: 20.26 | Improved |
| add selection function phi on queue (ablation) | SNOPS (NP) · SemanticPOSS | novel mIoU: 20.26 -> 20.63 | Improved |
| add tau_c on pseudo-label features (ablation) | SNOPS (NP) · SemanticPOSS | novel mIoU: 20.26 -> 20.90 | Improved |
| add both tau_c and phi (ablation) | SNOPS (NP) · SemanticPOSS | novel mIoU: 20.26 -> 21.40 | Improved |
| add semantic alignment branch (ablation) | SNOPS (NP+++) · SemanticPOSS | novel mIoU: 21.40 -> 30.05 | Improved |
Closest papers by meaning
- Novel Class Discovery for 3D Point Cloud Semantic Segmentation (CVPR 2023 · similarity 0.98)
- Novel Class Discovery for Point Cloud Segmentation via Joint Learning of Causal Representation and Reasoning (NeurIPS 2025 · similarity 0.91)
- Geometry and Uncertainty-Aware 3D Point Cloud Class-Incremental Semantic Segmentation (CVPR 2023 · similarity 0.91)
- PDF: A Probability-Driven Framework for Open World 3D Point Cloud Semantic Segmentation (CVPR 2024 · similarity 0.88)
- LogoSP: Local-global Grouping of Superpoints for Unsupervised Semantic Segmentation of 3D Point Clouds (CVPR 2025 · similarity 0.88)
- Rethinking Few-shot 3D Point Cloud Semantic Segmentation (CVPR 2024 · 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.
