Research map / 3D vision / Face and shape
GenSP: Consistent Spherical Parameterization via Learning Shape Generative Models
Sai Karthikey Pentapati, Shashank Gupta, Rajesh Sureddi, Yuezhi Yang, Alan Bovik, Qixing Huang
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 data augmentation with intermediate shapes bridging sphere and inputs (ablation) | GenSP · ShapeNet genus-0 test (100 shapes) | mean isometric distortion: 2.326 -> 1.681 | Improved |
| add data augmentation with intermediate shapes (ablation) | GenSP · ShapeNet genus-0 test (100 shapes) | max isometric distortion: 614.14 -> 7.719 | Improved |
| add data augmentation with intermediate shapes (ablation) | GenSP · ShapeNet genus-0 test (100 shapes) | consistency c(mi,mj): 3.164 -> 1.425 | Improved |
| add spanning-tree path correspondence propagation instead of latent interpolation (ablation) | GenSP · ShapeNet genus-0 test (100 shapes) | mean isometric distortion: 1.866 -> 1.681 | Improved |
| add spanning-tree path correspondence propagation (ablation) | GenSP · ShapeNet genus-0 test (100 shapes) | max isometric distortion: 23.228 -> 7.719 | Improved |
| add spanning-tree path correspondence propagation (ablation) | GenSP · ShapeNet genus-0 test (100 shapes) | consistency c(mi,mj): 1.613 -> 1.425 | Mixed |
| add geometric regularization losses (ablation) | GenSP · ShapeNet genus-0 test (100 shapes) | mean isometric distortion: 2.296 -> 1.681 | Improved |
| add geometric regularization losses (ablation) | GenSP · ShapeNet genus-0 test (100 shapes) | max isometric distortion: 271.26 -> 7.719 | Improved |
| add geometric regularization losses (ablation) | GenSP · ShapeNet genus-0 test (100 shapes) | consistency c(mi,mj): 1.740 -> 1.425 | Improved |
| use GenSP instead of best optimization baseline SMAT for consistency | SMAT · ShapeNet genus-0 test (100 shapes) | consistency c(mi,mj): 5.122 -> 1.425 | Improved |
| use GenSP instead of CMCF for distortion | CMCF · ShapeNet genus-0 test (100 shapes) | mean isometric distortion: 24.22 -> 1.681 | Improved |
| use GenSP instead of CMCF for conformal distortion | CMCF · ShapeNet genus-0 test (100 shapes) | mean conformal distortion: 1.029 -> 1.212 | Hurt |
Closest papers by meaning
- GenCorres: Consistent Shape Matching via Coupled Implicit-Explicit Shape Generative Models (ICLR 2024 · similarity 0.87)
- A Papier-Mâché Approach to Learning 3D Surface Generation (CVPR 2018 · similarity 0.87)
- Parameterization-Based Dataset Distillation of 3D Point Clouds through Learnable Shape Morphing (ICLR 2026 · similarity 0.87)
- SurfNet: Generating 3D Shape Surfaces Using Deep Residual Networks (CVPR 2017 · similarity 0.87)
- Dynamic Neural Surfaces for Elastic 4D Shape Representation and Analysis (CVPR 2025 · similarity 0.86)
- Flatten Anything: Unsupervised Neural Surface Parameterization (NeurIPS 2024 · 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.
