Research map / 3D vision / Face and shape
A Repulsive Force Unit for Garment Collision Handling in Neural Networks
Qingyang Tan, Yi Zhou, Tuanfeng Wang, Duygu Ceylan, Xin Sun, Dinesh Manocha
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 ReFU layer (Acc. SDF) fine-tuning | TailorNet · TailorNet garments (all 4 types) | CFMP: 49.09% -> 76.77% | Improved |
| add ReFU layer (Approx. SDF) | TailorNet · TailorNet garments (all 4 types) | CFMP: 49.09% -> 58.52% | Improved |
| add ReFU layer (Hybrid, accurate SDF for loss only) | TailorNet · TailorNet garments (all 4 types) | CFMP: 49.09% -> 68.71% | Improved |
| add ReFU layer (Acc. SDF) | TailorNet · TailorNet garments (all 4 types) | MPVE: 8.89 -> 8.64 | Improved |
| add ReFU layer (Acc. SDF) | TailorNet · TailorNet garments (all 4 types) | VFCP: 0.6% -> 0% | Improved |
| add ReFU layer (Acc. SDF) | TailorNet Shirt Male · TailorNet Shirt Male | CFMP: 11.92% -> 63.06% | Improved |
| use predicted scale instead of fixed scale alpha=1 (Acc. SDF) (ablation) | ReFU · TailorNet Shirt Male | CFMP: 37.22% -> 63.06% | Improved |
| use predicted scale instead of fixed scale alpha=1 (Approx. SDF) (ablation) | ReFU · TailorNet Shirt Male | CFMP: 4.88% -> 26.9% | Improved |
| use predicted scale instead of fixed scale (Approx. SDF) (ablation) | ReFU · TailorNet Shirt Male | Avg. VF collisions: 287.85 -> 110.12 | Improved |
| per-vertex latent k(z)_i instead of shared latent (Alt. 1) (ablation) | ReFU (Approx. SDF) · TailorNet Shirt Male | CFMP: 21.36% -> 26.9% | Improved |
| use global latent plus SDF instead of SDF only (Alt. 2) (ablation) | ReFU (Approx. SDF) · TailorNet Shirt Male | CFMP: 20.81% -> 26.9% | Improved |
| naive post-processing with approximate SDF | TailorNet · TailorNet garments (all 4 types) | CFMP: 49.09% -> 30.89% | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- TailorNet — TailorNet: Predicting Clothing in 3D as a Function of Human Pose, Shape and Garment Style (CVPR 2020) · better on 3 of 3 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
- Self-Supervised Collision Handling via Generative 3D Garment Models for Virtual Try-On (CVPR 2021 · similarity 0.89)
- PhysDrape: Learning Explicit Forces and Collision Constraints for Physically Realistic Garment Draping (ECCV 2026 · similarity 0.88)
- SNUG: Self-Supervised Neural Dynamic Garments (CVPR 2022 · similarity 0.87)
- Learning 3D Garment Animation from Trajectories of A Piece of Cloth (NeurIPS 2024 · similarity 0.86)
- GarmentLab: A Unified Simulation and Benchmark for Garment Manipulation (NeurIPS 2024 · similarity 0.86)
- DexGarmentLab: Dexterous Garment Manipulation Environment with Generalizable Policy (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.
