Research map / 3D vision / 3D generation
Fine Detailed Texture Learning for 3D Meshes With Generative Models
Ayşegül Dündar, Jun Gao, Andrew Tao, Bryan Catanzaro
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 |
|---|---|---|---|
| add learnable position embedding to the discriminator input (ablation) | GAN texture pipeline · Tripod | FID: 82.19 -> 66.57 | Improved |
| add position-based attention to the generator (ablation) | GAN texture pipeline with discriminator embedding · Tripod | FID: 66.57 -> 64.55 | Improved |
| mask pseudo ground-truth with 3D projected silhouettes instead of Mask-RCNN masks (ablation) | GAN texture pipeline · Tripod | FID: 84.90 -> 82.19 | Improved |
| prune frames with wrongly estimated camera parameters (ablation) | GAN texture pipeline · Tripod | FID: 100.43 -> 82.19 | Improved |
| replace position-based attention with self-attention (ablation) | generator · Tripod | FID | No clear effect |
| add a second attention layer in the generator (ablation) | generator · Tripod | FID | No clear effect |
| add learned embedding in discriminator (ablation) | ConvMesh · Pascal 3D+ car | avg. FID: 29.06 -> 26.95 | Improved |
| add generator attention (ablation) | ConvMesh with discriminator embedding · Pascal 3D+ car | avg. FID: 26.95 -> 26.79 | Improved |
| add learned embedding in discriminator (ablation) | ConvMesh · CUB | avg. FID: 40.80 -> 39.47 | Improved |
| add generator attention (ablation) | ConvMesh with discriminator embedding · CUB | avg. FID: 39.47 -> 38.45 | Improved |
Closest papers by meaning
- Learning Generative Models of Textured 3D Meshes From Real-World Images (ICCV 2021 · similarity 0.93)
- Convolutional Generation of Textured 3D Meshes (NeurIPS 2020 · similarity 0.92)
- Progressive Learning of 3D Reconstruction Network From 2D GAN Data (TPAMI 2023 · similarity 0.91)
- Leveraging 2D Data to Learn Textured 3D Mesh Generation (CVPR 2020 · similarity 0.91)
- GET3D: A Generative Model of High Quality 3D Textured Shapes Learned from Images (NeurIPS 2022 · similarity 0.90)
- AUV-Net: Learning Aligned UV Maps for Texture Transfer and Synthesis (CVPR 2022 · similarity 0.90)
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.
