Research map / Label-efficient and robust learning / Self-supervised and contrastive
SENC: Handling Self-collision in Neural Cloth Simulation
Zhouyingcheng Liao, Sinan Wang, Taku Komura
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 penetration-volume self-collision loss (GIA) and self-collision edges to HOOD | HOOD · AMASS test, t-shirt | self-collision loss (x1e-3): 26.10 -> 1.80 | Improved |
| add self-collision loss and edges to HOOD | HOOD · AMASS test, skirt | self-collision loss (x1e-3): 7.33 -> 1.59 | Improved |
| use penetration volume instead of penetration surface area for the loss (ablation) | SENC · AMASS test, t-shirt | self-collision loss (x1e-3): 28.81 -> 1.80 | Improved |
| add spatial self-collision edges to the GNN (ablation) | SENC · AMASS test, skirt | self-collision loss (x1e-3): 7.92 -> 1.59 | Improved |
| add spatial self-collision edges to the GNN (ablation) | SENC · AMASS test, t-shirt | self-collision loss (x1e-3): 1.14 -> 1.80 | No clear effect |
| exclude original mesh edges when building self-collision edges (ablation) | SENC · AMASS test | self-collision loss and speed: FPS average 16.196 -> 16.843; skirt loss 2.02 -> 1.59 (t-shirt 2.37 -> 1.80) | Improved |
| use penetration-volume loss instead of repulsive loss (ablation) | HOOD · AMASS test, t-shirt | self-collision loss (x1e-3): 15.16 -> 1.80 | Improved |
| add self-collision edges and GIA loss | HOOD · HOOD test garments | self-collision loss: tshirt dynamic 164.23 -> 6.58 | Improved |
| add self-collision edges and computation | HOOD · runtime | FPS: 19.090 -> 16.843 | Hurt |
Later papers that compared with this method
2 later accepted papers report SENC as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- Self-supervised Garment Dynamics with Persistent Wrinkles (ECCV 2026 · better on 2 of 2 shared results)
- MUSE: Multi-Scale Dense Self-Distillation for Nucleus Detection and Classification (AAAI 2026 · better on 1 of 1 shared result)
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
- Differentiable Cloth Simulation for Inverse Problems (NeurIPS 2019 · similarity 0.88)
- NeuralClothSim: Neural Deformation Fields Meet the Thin Shell Theory (NeurIPS 2024 · similarity 0.86)
- GarmentLab: A Unified Simulation and Benchmark for Garment Manipulation (NeurIPS 2024 · similarity 0.86)
- Learning 3D Garment Animation from Trajectories of A Piece of Cloth (NeurIPS 2024 · similarity 0.86)
- DexGarmentLab: Dexterous Garment Manipulation Environment with Generalizable Policy (NeurIPS 2025 · similarity 0.84)
- GarmentPile: Point-Level Visual Affordance Guided Retrieval and Adaptation for Cluttered Garments Manipulation (CVPR 2025 · similarity 0.84)
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.
