Research map / 3D vision / Face and shape
GET3D: A Generative Model of High Quality 3D Textured Shapes Learned from Images
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 GET3D rather than EG3D for explicit 3D textured shape generation | ShapeNet Car · ShapeNet Car | FID-3D: 21.89 -> 10.25 | Helped |
| Use GET3D rather than EG3D for explicit 3D textured shape generation | ShapeNet Chair · ShapeNet Chair | FID-3D: 46.06 -> 23.28 | Helped |
| Use GET3D rather than EG3D for explicit 3D textured shape generation | ShapeNet Motorbike · ShapeNet Motorbike | FID-3D: 89.97 -> 65.60 | Helped |
| Use GET3D rather than EG3D for explicit 3D textured shape generation | TurboSquid Animal · TurboSquid Animal | FID-3D: 83.47 -> 28.33 | Helped |
| Increase training image resolution from 128² to 1024² (ablation) | GET3D · ShapeNet Car | FID: 39.21 -> 10.25 | Helped |
| Increase training image resolution from 128² to 1024² (ablation) | GET3D · ShapeNet Chair | FID: 43.04 -> 23.28 | Helped |
| Add volume subdivision (ablation) | GET3D · ShapeNet Motorbike | FID: 65.60 -> 54.12 | Helped |
| Add volume subdivision (ablation) | GET3D · ShapeNet Chair | FID: 23.28 -> 23.17 | No clear effect |
| Remove camera-pose conditioning from the discriminator (ablation) | GET3D · ShapeNet Car | FID: 10.25 -> 11.63 | Hurt |
| Add Gaussian noise to camera poses during training (ablation) | GET3D · ShapeNet Car | FID: 10.25 -> 19.53 | Hurt |
| Replace ground-truth masks with predicted Mask-Black silhouettes (ablation) | GET3D · ShapeNet Car | FID: 10.25 -> 29.68 | Hurt |
| Replace ground-truth masks with predicted Mask-Random silhouettes (ablation) | GET3D · ShapeNet Car | FID: 10.25 -> 33.16 | Hurt |
Closest papers by meaning
- MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers (ICLR 2025 · similarity 0.89)
- Magic123: One Image to High-Quality 3D Object Generation Using Both 2D and 3D Diffusion Priors (ICLR 2024 · similarity 0.89)
- Structured 3D Latents for Scalable and Versatile 3D Generation (CVPR 2025 · similarity 0.89)
- En3D: An Enhanced Generative Model for Sculpting 3D Humans from 2D Synthetic Data (CVPR 2024 · similarity 0.88)
- Rigel3D: Rig-aware Latents for Animation-Ready 3D Asset Generation (NeurIPS 2026 · similarity 0.88)
- MeshGen: Generating PBR Textured Mesh with Render-Enhanced Auto-Encoder and Generative Data Augmentation (CVPR 2025 · similarity 0.88)
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.