Research map / Point clouds / Shape and surface completion
GarNet: A Two-Stream Network for Fast and Accurate 3D Cloth Draping
Erhan Gundogdu, Victor Constantin, Amrollah Seifoddini, Minh Dang, Mathieu Salzmann, Pascal Fua
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 patch-wise garment features (Local or Global instead of late fusion) (ablation) | GarNet · T-shirt draping (own dataset) | E_dist (cm): 1.78 -> 0.93 | Improved |
| add patch-wise garment features (Local instead of late fusion) (ablation) | GarNet · Jeans draping (own dataset) | E_dist (cm): 2.13 -> 0.88 | Improved |
| add patch-wise garment features (Local instead of late fusion) (ablation) | GarNet · Sweater draping (own dataset) | E_dist (cm): 1.13 -> 0.97 | Improved |
| use nearest-neighbor pooled local body features (Local instead of Global) (ablation) | GarNet · Jeans draping (own dataset) | E_dist (cm): 1.01 -> 0.88 | Improved |
| use nearest-neighbor pooled local body features (Local instead of Global) (ablation) | GarNet · T-shirt draping (own dataset) | E_dist (cm): 1.05 -> 0.93 | Improved |
| use nearest-neighbor pooled local body features (Local instead of Global) (ablation) | GarNet · Wang18 shirt dataset | normalized L2 distance (%): 1.15 -> 0.89 | Improved |
| GarNet-Local versus DQS skinning only | T-shirt draping · T-shirt draping (own dataset) | E_dist (cm): 9.98 -> 0.93 | Improved |
| GarNet-Local versus prior learned method Wang18 | shirt draping · Wang18 shirt dataset | normalized L2 distance (%): 3.01 -> 0.89 | Improved |
| GarNet-Local versus physics-based simulation in speed | garment draping · Garment draping (own dataset) | computation time (ms): > 19000 -> 68 | Improved |
| reduce training poses from 50% to 25% of poses (ablation) | GarNet-Local · T-shirt unseen poses | E_dist (cm): 1.16 -> 1.68 | Hurt |
| add normal loss term and bending term to vertex+penetration loss (ablation) | GarNet loss · Wang18 shirt dataset (loss ablation) | E_normal (deg): 8.88 -> 7.36 | Improved |
| add bending term to vertex+penetration+normal loss (ablation) | GarNet loss · Wang18 shirt dataset (loss ablation) | E_dist: 1.08 -> 0.72 | Improved |
Closest papers by meaning
- GarNet++: Improving Fast and Accurate Static 3D Cloth Draping by Curvature Loss (TPAMI 2020 · similarity 0.95)
- GarmentCodeData: A Dataset of 3D Made-to-Measure Garments With Sewing Patterns (ECCV 2024 · similarity 0.90)
- DeepCloth: Neural Garment Representation for Shape and Style Editing (TPAMI 2022 · similarity 0.90)
- Garment4D: Garment Reconstruction from Point Cloud Sequences (NeurIPS 2021 · similarity 0.90)
- Deep Fashion3D: A Dataset and Benchmark for 3D Garment Reconstruction from Single Images (ECCV 2020 · similarity 0.90)
- Learning 3D Garment Animation from Trajectories of A Piece of Cloth (NeurIPS 2024 · similarity 0.89)
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.
