Research map / 3D vision / Face and shape
Convolutional Generation of Textured 3D Meshes
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 |
|---|---|---|---|
| add sinusoidal positional encoding to discriminator input (ablation) | class-conditioned mesh GAN · CUB 512x512 | truncated FID: 43.71 -> 33.63 | Helped |
| update D twice per G update instead of once (ablation) | class-conditioned mesh GAN · CUB 512x512 | truncated FID: 41.55 -> 33.63 | Helped |
| add instance normalization in discriminator (ablation) | class-conditioned mesh GAN · CUB 512x512 | truncated FID: 33.63 -> 36.38 | Hurt |
| use attention over RNN states for text conditioning instead of concatenated sentence vector (ablation) | text-conditioned mesh GAN · CUB 512x512 | truncated FID: 22.14 -> 18.45 | Helped |
| add class label conditioning vs unconditional | mesh GAN · CUB 512x512 | truncated FID: 41.56 -> 33.63 | Helped |
| add text conditioning vs class conditioning | mesh GAN · CUB 512x512 | truncated FID: 33.63 -> 18.45 | Helped |
| add color attribute to class conditioning | mesh GAN · P3D 512x512 | truncated FID: 27.73 -> 31.30 | Hurt |
| increase texture resolution 256 to 512 | mesh GAN class+color · P3D | truncated FID: 39.09 -> 31.30 | Helped |
| increase texture resolution 256 to 512 (class conditioning) | mesh GAN · CUB | truncated FID: 33.55 -> 33.63 | No clear effect |
Closest papers by meaning
- Magic123: One Image to High-Quality 3D Object Generation Using Both 2D and 3D Diffusion Priors (ICLR 2024 · similarity 0.89)
- Paint-it: Text-to-Texture Synthesis via Deep Convolutional Texture Map Optimization and Physically-Based Rendering (CVPR 2024 · similarity 0.89)
- MeshGen: Generating PBR Textured Mesh with Render-Enhanced Auto-Encoder and Generative Data Augmentation (CVPR 2025 · similarity 0.89)
- MeshGPT: Generating Triangle Meshes with Decoder-Only Transformers (CVPR 2024 · similarity 0.89)
- One-2-3-45++: Fast Single Image to 3D Objects with Consistent Multi-View Generation and 3D Diffusion (CVPR 2024 · similarity 0.89)
- Single Mesh Diffusion Models with Field Latents for Texture Generation (CVPR 2024 · 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.