Research map / 3D vision / Face and shape
DrapeNet: Garment Generation and Self-Supervised Draping
Luca De Luigi, Ren Li, Benoît Guillard, Mathieu Salzmann, Pascal Fua
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 |
|---|---|---|---|
| Use self-supervised physics-based draping with a garment-conditioned refinement field | DrapeNet · CLOTH3D test top garments | B2G-top (%): 7.2 -> 0.9 | Improved |
| Use self-supervised physics-based draping with a garment-conditioned refinement field | DrapeNet · CLOTH3D test bottom garments | B2G-bottom (%): 3.4 -> 0.3 | Improved |
| Use the Intersection Solver (ablation) | DrapeNet · CLOTH3D test top-bottom garment pairs | G2G (%): 1.1 -> 0.5 | Improved |
| Use physics-based self-supervision rather than matching the simulator's exact vertex locations | DrapeNet · CLOTH3D test top garments | ED-top (mm): 28.1 -> 47.9 | Hurt |
| Use physics-based self-supervision rather than matching the simulator's exact vertex locations | DrapeNet · CLOTH3D test bottom garments | ED-bottom (mm): 18.3 -> 27.3 | Hurt |
| Add gradient supervision to the garment UDF training loss (ablation) | DrapeNet garment generative network · Garment reconstruction test examples | surface quality | Improved |
| Use BCE instead of L1 for the distance loss, with gradient supervision (ablation) | DrapeNet garment generative network · Garment reconstruction test examples | surface quality | Improved |
| Use BCE instead of L2 for the distance loss, with gradient supervision (ablation) | DrapeNet garment generative network · Garment reconstruction test examples | holes in reconstructed surfaces | Improved |
| Add a waist-and-hip pin loss for bottom garments (ablation) | DrapeNet bottom draping network · Bottom garment draping | deformation naturalness | Improved |
| Set the pin-loss λ to 0 or 1 instead of 0.1 (ablation) | DrapeNet bottom draping network · Bottom garment draping | deformation naturalness | Hurt |
| Add the layer loss to enforce top-bottom garment ordering (ablation) | DrapeNet · Top-bottom garment draping | garment intersections | Improved |
| Condition the refinement field on the correct garment latent code (ablation) | DrapeNet · Garment draping with mismatched latent codes | deformation realism | Improved |
Later papers that compared with this method
5 later accepted papers report DrapeNet as a baseline in their results tables; 5 of them report a better number on most of the results they share with it.
- PhysDrape: Learning Explicit Forces and Collision Constraints for Physically Realistic Garment Draping (ECCV 2026 · better on 3 of 3 shared results)
- Design2Cloth: 3D Cloth Generation from 2D Masks (CVPR 2024 · better on 6 of 6 shared results)
- Garment Recovery with Shape and Deformation Priors (CVPR 2024 · better on 3 of 3 shared results)
- WordRobe: Text-Guided Generation of Textured 3D Garments (ECCV 2024 · better on 2 of 2 shared results)
- ISP: Multi-Layered Garment Draping with Implicit Sewing Patterns (NeurIPS 2023 · better on 1 of 1 shared result)
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.
- DeePSD — DeePSD: Automatic Deep Skinning and Pose Space Deformation for 3D Garment Animation (ICCV 2021) · better on 3 of 5 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
- Multi-Garment Net: Learning to Dress 3D People From Images (ICCV 2019 · similarity 0.90)
- SNUG: Self-Supervised Neural Dynamic Garments (CVPR 2022 · similarity 0.90)
- GarNet: A Two-Stream Network for Fast and Accurate 3D Cloth Draping (ICCV 2019 · similarity 0.90)
- Self-Supervised Collision Handling via Generative 3D Garment Models for Virtual Try-On (CVPR 2021 · similarity 0.89)
- Design2Cloth: 3D Cloth Generation from 2D Masks (CVPR 2024 · similarity 0.88)
- Learning 3D Garment Animation from Trajectories of A Piece of Cloth (NeurIPS 2024 · 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.
