Research map / 3D vision / Face and shape
3D Shape Generation With Grid-Based Implicit Functions
Check 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 |
|---|---|---|---|
| Generate local grid-based implicit representations instead of a global implicit representation | 3D shape GAN generation · ShapeNet Core v1, Avg. | COV(%): 73.36 -> 82.80 | Helped |
| Generate local grid-based implicit representations instead of a global implicit representation | 3D shape GAN generation · ShapeNet Core v1, Avg. | MMD: 2,831 -> 2,607 | Helped |
| Generate local grid-based implicit representations instead of a global implicit representation | 3D shape GAN generation · ShapeNet Core v1, Chair | ECD: 2,553 -> 144 | Helped |
| Replace patch discriminators with a standard discriminator (ablation) | Our grid-based GAN · ShapeNet Core v1, Chair | COV: 82.09 -> 0.00 | Hurt |
| Add patch discriminators at resolutions 16 and 8 (ablation) | Our grid-based GAN · ShapeNet Core v1, Chair | COV: 0.00 -> 74.71 | Helped |
| Add the resolution-32 patch discriminator to the 16 and 8 discriminators (ablation) | Our grid-based GAN · ShapeNet Core v1, Chair | COV: 74.71 -> 82.09 | Helped |
| Add the resolution-16 patch discriminator to the 32 and 8 discriminators (ablation) | Our grid-based GAN · ShapeNet Core v1, Chair | COV: 76.40 -> 82.09 | Helped |
| Add the resolution-8 patch discriminator to the 32 and 16 discriminators (ablation) | Our grid-based GAN · ShapeNet Core v1, Chair | COV: 81.49 -> 82.09 | Helped |
Closest papers by meaning
- 3DILG: Irregular Latent Grids for 3D Generative Modeling (NeurIPS 2022 · similarity 0.90)
- Learning Representations and Generative Models for 3D Point Clouds (ICML 2018 · similarity 0.89)
- Rank3DGAN: Semantic Mesh Generation Using Relative Attributes (AAAI 2020 · similarity 0.89)
- Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling (NeurIPS 2016 · similarity 0.88)
- Unsupervised 3D Shape Completion Through GAN Inversion (CVPR 2021 · similarity 0.88)
- What You See is What You GAN: Rendering Every Pixel for High-Fidelity Geometry in 3D GANs (CVPR 2024 · similarity 0.86)
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.