Research map / Detection and segmentation / 3D detection and driving
PartNet: A Large-Scale Benchmark for Fine-Grained and Hierarchical Part-Level 3D Object Understanding
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 detection-by-segmentation with jointly predicted disjoint instance masks and confidence scores instead of SGPN's similarity-based grouping | PointNet++ for PartNet instance segmentation · PartNet | part-category mAP (%): 46.8 -> 54.4 | Helped |
| Use ensemble inference across segmentation levels rather than bottom-up inference | PointNet++ hierarchical semantic segmentation · PartNet | part-category mIoU (%): 51.2 -> 51.7 | Helped |
| Use ensemble inference across segmentation levels rather than top-down inference | PointNet++ hierarchical semantic segmentation · PartNet | part-category mIoU (%): 50.8 -> 51.7 | Helped |
| Refine hierarchical templates using multiple annotations to merge confusing concepts or remove hard-to-distinguish parts | PartNet annotation templates · PartNet multiple-annotation subset | annotation consistency (%): 69.8 -> 83.3; 13.5% improvement | Helped |
| Use PointCNN rather than PointNet++ | Fine-grained semantic segmentation · PartNet | part-category mIoU (%): 58.1 -> 59.8 | Helped |
| Use PointNet++ rather than PointCNN | Fine-grained semantic segmentation · PartNet | shape mIoU (%): 68.1 -> 66.3 | Helped |
Closest papers by meaning
- PartNeXt: A Next-Generation Dataset for Fine-Grained and Hierarchical 3D Part Understanding (NeurIPS 2025 · similarity 0.94)
- 3DCoMPaT200: Language Grounded Large-Scale 3D Vision Dataset for Compositional Recognition (NeurIPS 2024 · similarity 0.90)
- PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers (NeurIPS 2025 · similarity 0.89)
- BFANet: Revisiting 3D Semantic Segmentation with Boundary Feature Analysis (CVPR 2025 · similarity 0.89)
- PartDistill: 3D Shape Part Segmentation by Vision-Language Model Distillation (CVPR 2024 · similarity 0.89)
- SUM Parts: Benchmarking Part-Level Semantic Segmentation of Urban Meshes (CVPR 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.