Research map / 3D vision / Face and shape
Coupling Explicit and Implicit Surface Representations for Generative 3D Modeling
Omid Poursaeed, Matthew Fisher, Noam Aigerman, Vladimir G. Kim
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 |
|---|---|---|---|
| couple explicit AtlasNet and implicit OccupancyNet with consistency losses (AN branch) | AtlasNet · Single-view reconstruction | Chamfer-L1: 1.74 -> 1.60 | Improved |
| couple explicit AtlasNet and implicit OccupancyNet (ON branch) | OccupancyNet · Single-view reconstruction | Chamfer-L1: 1.92 -> 1.65 | Improved |
| hybrid training (AN branch) | AtlasNet · Single-view reconstruction | Normal consistency: 81.8 -> 84.9 | Improved |
| hybrid training (AN branch) | AtlasNet · Point cloud auto-encoding | Chamfer-L1: 0.58 -> 0.46 | Improved |
| hybrid training (ON branch) | OccupancyNet · Point cloud auto-encoding | Chamfer-L1: 0.45 -> 0.41 | Improved |
| hybrid training (AN branch) | AtlasNet · Point cloud auto-encoding | Normal consistency: 85.4 -> 88.3 | Improved |
| hybrid training (ON branch) | OccupancyNet · Point cloud auto-encoding | Normal consistency: 89.8 -> 90.2 | Improved |
| add image loss L_img (ablation) | Hybrid AN · Single-view reconstruction | Chamfer-L1: 1.64 -> 1.60 | Improved |
| add image loss L_img (ablation) | Hybrid AN · Point cloud auto-encoding | Chamfer-L1: 0.49 -> 0.46 | Improved |
| add normal consistency loss L_norm (ablation) | Hybrid AN · Single-view reconstruction | Normal consistency: 83.9 -> 84.9 | Improved |
| add normal consistency loss L_norm (ablation) | Hybrid AN · Point cloud auto-encoding | Normal consistency: 87.6 -> 88.3 | Improved |
| add level-set consistency loss alone (no L_img, L_norm) (ablation) | Hybrid AN · Single-view reconstruction | Chamfer-L1: 1.74 -> 1.68 | Improved |
Closest papers by meaning
- Iso-Points: Optimizing Neural Implicit Surfaces With Hybrid Representations (CVPR 2021 · similarity 0.91)
- Learning to Infer Implicit Surfaces without 3D Supervision (NeurIPS 2019 · similarity 0.90)
- A Papier-Mâché Approach to Learning 3D Surface Generation (CVPR 2018 · similarity 0.90)
- Implicit Surface Representations As Layers in Neural Networks (ICCV 2019 · similarity 0.90)
- Neural Vector Fields: Implicit Representation by Explicit Learning (CVPR 2023 · similarity 0.90)
- Deep Implicit Surface Point Prediction Networks (ICCV 2021 · 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.
