Research map / Label-efficient and robust learning / Recognition and long-tail
Deep Shells: Unsupervised Shape Correspondence with Optimal Transport
Read the paperCheck 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 |
|---|---|---|---|
| train jointly on FAUST+SCAPE including inter-dataset pairs instead of individual dataset | Deep Shells · FAUST test | mean geodesic error (% diameter): 1.7 -> 1.6 | Helped |
| train jointly on FAUST+SCAPE including inter-dataset pairs instead of individual dataset | Deep Shells · SCAPE test | mean geodesic error (% diameter): 2.5 -> 2.4 | Helped |
| replace stochastic MCMC pose search in Smooth Shells with learned features in Deep Shells | Smooth Shells · FAUST remeshed | mean geodesic error (% diameter): 2.5 -> 1.7 | Helped |
| replace stochastic MCMC pose search in Smooth Shells with learned features in Deep Shells | Smooth Shells · SCAPE remeshed | mean geodesic error (% diameter): 4.7 -> 2.5 | Helped |
| replace Smooth Shells pipeline with Deep Shells for test-time total cost | Smooth Shells · test-time total runtime | total runtime (s): 125. -> 14. | Helped |
| train on only 10 shapes versus GeoFMNet baseline | Deep Shells · SURREAL train, FAUST test | mean geodesic error | Helped |
Closest papers by meaning
- Unsupervised Learning of Dense Shape Correspondence (CVPR 2019 · similarity 0.90)
- NCP: Neural Correspondence Prior for Effective Unsupervised Shape Matching (NeurIPS 2022 · similarity 0.90)
- NeuroMorph: Unsupervised Shape Interpolation and Correspondence in One Go (CVPR 2021 · similarity 0.88)
- G-MSM: Unsupervised Multi-Shape Matching With Graph-Based Affinity Priors (CVPR 2023 · similarity 0.88)
- Integrating Efficient Optimal Transport and Functional Maps For Unsupervised Shape Correspondence Learning (CVPR 2024 · similarity 0.88)
- DcMatch: Unsupervised Multi-Shape Matching with Dual-Level Consistency (AAAI 2026 · 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.