Research map / 3D vision / Face and shape
SyncSpecCNN: Synchronized Spectral CNN for 3D Shape Segmentation
Li Yi, Hao Su, Xingwen Guo, Leonidas J. Guibas
Check 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 SpecTN to synchronize spectral domains (ablation) | SyncSpecCNN · Cross-category part segmentation | IoU: 79.65 -> 81.97 | Improved |
| Add SpecTN to synchronize spectral domains (ablation) | SyncSpecCNN · Cross-category part segmentation | classification accuracy: 99.59 -> 99.71 | Improved |
| Add SpecTN to synchronize spectral domains (ablation) | SyncSpecCNN · Partial-shape part segmentation | IoU: 76.19 -> 78.02 | Improved |
| Use modulated exponential window instead of exponential window kernels (ablation) | SyncSpecCNN with small kernels · Cross-category part segmentation | IoU: 0.6285 -> 0.6997 | Improved |
| Use modulated exponential window instead of exponential window kernels (ablation) | SyncSpecCNN with large kernels · Cross-category part segmentation | IoU: 0.7223 -> 0.7341 | Improved |
| Use modulated exponential window instead of exponential window kernels (ablation) | SyncSpecCNN with multiscale kernels · Cross-category part segmentation | IoU: 0.7386 -> 0.7524 | Improved |
| Use multiscale kernels instead of small kernels alone (ablation) | SyncSpecCNN with exponential window basis · Cross-category part segmentation | IoU: 0.6285 -> 0.7386 | Improved |
| Use multiscale kernels instead of large kernels alone (ablation) | SyncSpecCNN with exponential window basis · Cross-category part segmentation | IoU: 0.7223 -> 0.7386 | Improved |
| Use multiscale kernels instead of small kernels alone (ablation) | SyncSpecCNN with modulated exponential window basis · Cross-category part segmentation | IoU: 0.6997 -> 0.7524 | Improved |
| Use multiscale kernels instead of large kernels alone (ablation) | SyncSpecCNN with modulated exponential window basis · Cross-category part segmentation | IoU: 0.7341 -> 0.7524 | Improved |
| Use modulated exponential window with small kernels instead of cubic spline basis (ablation) | SyncSpecCNN · Cross-category part segmentation | IoU: 0.5369 -> 0.6997 | Improved |
Later papers that compared with this method
2 later accepted papers report SyncSpecCNN as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.
- Laplacian Mesh Transformer: Dual Attention and Topology Aware Network for 3D Mesh Classification and Segmentation (ECCV 2022 · better on 0 of 1 shared result)
- Learning Part Generation and Assembly for Structure-Aware Shape Synthesis (AAAI 2020 · better on 1 of 2 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- FeaStNet: Feature-Steered Graph Convolutions for 3D Shape Analysis (CVPR 2018 · similarity 0.88)
- Spectral Shape Recovery and Analysis Via Data-driven Connections (IJCV 2021 · similarity 0.87)
- Local Spectral Graph Convolution for Point Set Feature Learning (ECCV 2018 · similarity 0.86)
- From Feature Learning to Spectral Basis Learning: A Unifying and Flexible Framework for Efficient and Robust Shape Matching (CVPR 2026 · similarity 0.86)
- Laplacian Mesh Transformer: Dual Attention and Topology Aware Network for 3D Mesh Classification and Segmentation (ECCV 2022 · similarity 0.86)
- SegGraph: Leveraging Graphs of SAM Segments for Few-Shot 3D Part Segmentation (NeurIPS 2025 · similarity 0.85)
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.
