Research map / Point clouds / Shape and surface completion
Point-Based Modeling of Human Clothing
Ilya Zakharkin, Kirill Mazur, Artur Grigorev, Victor Lempitsky
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 |
|---|---|---|---|
| use point-cloud outfit model instead of Tex2Shape | single-image clothing geometry · AzurePeople user study | user preference: 65.6% vs 34.3% | Improved |
| use point-cloud outfit model instead of Tex2Shape | single-image clothing geometry · PeopleSnapshot user study | user preference: 38.1% vs 61.9% | Hurt |
| use point-cloud outfit model instead of MGN | single-image clothing geometry · AzurePeople user study | user preference: 74.5% vs 25.5% | Improved |
| use point-cloud outfit model instead of MGN | single-image clothing geometry · PeopleSnapshot user study | user preference: 50.9% vs 49.1% | No clear effect |
| use point-cloud outfit model instead of Octopus | single-image clothing geometry · AzurePeople user study | user preference: 73.7% vs 26.3% | Improved |
| use point-cloud outfit model instead of Octopus | single-image clothing geometry · PeopleSnapshot user study | user preference: 47.8% vs 52.2% | No clear effect |
| use point-cloud outfit model instead of MGN | geometry fitting from masks · BCNet | Chamfer distance: 0.0025 -> 0.00121 | Improved |
| use point-cloud geometry scaffold instead of SMPL-X neural texture (StylePeople) | appearance retargeting · PeopleSnapshot | LPIPS: 0.0569 -> 0.031 | Improved |
| use point-cloud geometry scaffold instead of SMPL-X neural texture (StylePeople) | appearance retargeting · AzurePeople | LPIPS: 0.0693 -> 0.066 | Improved |
| use point-cloud geometry scaffold instead of SMPL-X neural texture (StylePeople) | appearance retargeting · AzurePeople | MS-SSIM: 0.946 -> 0.937 | Hurt |
| use point-cloud geometry scaffold instead of SMPL-X neural texture (StylePeople) | appearance retargeting · PeopleSnapshot | SSIM: 0.938 -> 0.950 | Improved |
Closest papers by meaning
- The Power of Points for Modeling Humans in Clothing (ICCV 2021 · similarity 0.95)
- CloSET: Modeling Clothed Humans on Continuous Surface With Explicit Template Decomposition (CVPR 2023 · similarity 0.91)
- SCALE: Modeling Clothed Humans with a Surface Codec of Articulated Local Elements (CVPR 2021 · similarity 0.91)
- Garment4D: Garment Reconstruction from Point Cloud Sequences (NeurIPS 2021 · similarity 0.90)
- SCULPT: Shape-Conditioned Unpaired Learning of Pose-dependent Clothed and Textured Human Meshes (CVPR 2024 · similarity 0.87)
- Garment Recovery with Shape and Deformation Priors (CVPR 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.
