Research map / Point clouds / Shape and surface completion
A Large-scale Annotated Mechanical Components Benchmark for Classification and Retrieval Tasks with Deep Neural Networks
Sangpil Kim, Hyung-gun Chi, Xiao Hu, Qixing Huang, Karthik Ramani
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 |
|---|---|---|---|
| move from aligned dataset A to unaligned dataset B | RotationNet · MCB retrieval, micro | mAP: 0.805 -> 0.607 | Hurt |
| move from aligned dataset A to unaligned dataset B | VRN · MCB retrieval, micro | mAP: 0.653 -> 0.628 | Hurt |
| move from aligned dataset A to unaligned dataset B | DLAN · MCB retrieval, micro | mAP: 0.879 -> 0.908 | Improved |
| move from aligned dataset A to unaligned dataset B | PointNet++ · MCB classification | accuracy over object (%): 87.45 -> 93.91 | Improved |
| move from aligned dataset A to unaligned dataset B | VRN · MCB classification | accuracy over object (%): 93.17 -> 85.44 | Hurt |
| move from ModelNet40 to mechanical components (dataset A) | MVCNN · classification | accuracy over object (%): 95.0 -> 64.67 | Hurt |
| move from ModelNet40 to mechanical components (dataset A) | RotationNet · classification | accuracy over object (%): 97.37 -> 97.35 | Hurt |
| move from ModelNet40 to mechanical components (dataset A) | DLAN · classification | accuracy over object (%): 84.0 -> 93.53 | Improved |
| add view-order correspondence estimation (RotationNet vs MVCNN) | RotationNet · MCB dataset A classification | accuracy over object (%): 64.67 -> 97.35 | Improved |
| increase point cloud density from 128 to 2,048 points (ablation) | PointNet++, SpiderCNN, PointCNN · MCB dataset B classification | accuracy | Improved |
Closest papers by meaning
- Survey and Evaluation of Neural 3D Shape Classification Approaches (TPAMI 2021 · similarity 0.86)
- 3DCoMPaT200: Language Grounded Large-Scale 3D Vision Dataset for Compositional Recognition (NeurIPS 2024 · similarity 0.85)
- PartNeXt: A Next-Generation Dataset for Fine-Grained and Hierarchical 3D Part Understanding (NeurIPS 2025 · similarity 0.84)
- Activating Sparse Part Concepts for 3D Class Incremental Learning (CVPR 2025 · similarity 0.83)
- Imagine: Image-Guided 3D Part Assembly with Structure Knowledge Graph (AAAI 2025 · similarity 0.83)
- Error Notebook-Guided, Training-Free Part Retrieval in 3D CAD Assemblies via Vision-Language Models (ICLR 2026 · similarity 0.83)
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.
