Research map / 3D vision / Face and shape
Adversarial Texture Optimization From RGB-D Scans
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 |
|---|---|---|---|
| use the conditional adversarial texture objective instead of L1 | RGB-D texture optimization · Synthetic 2D example | nearest patch loss: 7.33 -> 1.53 | Helped |
| use the conditional adversarial texture objective instead of L1 | RGB-D texture optimization · Synthetic 2D example | exact patch loss: 10.7 -> 11.3 | Hurt |
| use adversarial texture optimization instead of the strongest baseline, 3DLite | RGB-D texture optimization · Real object scans | perceptual loss: 0.185 -> 0.175 | Helped |
| use adversarial texture optimization instead of the strongest baseline, 3DLite | RGB-D texture optimization · ScanNet | perceptual loss: 0.445 -> 0.395 | Helped |
| use adversarial texture optimization instead of the strongest baseline, L1 | RGB-D texture optimization · CAD models | perceptual loss: 0.199 -> 0.176 | Helped |
| remove the auxiliary view from discriminator training (ablation) | conditional adversarial texture optimization · misalignment-tolerant discriminator evaluation | misalignment tolerance | Hurt |
| remove the condition from the discriminator (ablation) | conditional adversarial texture optimization · local patch evaluation | local patch disambiguation | Hurt |
| train the discriminator with more than 25 input frames (ablation) | conditional adversarial texture optimization · sparse-view scans | texture quality | Helped |
Closest papers by meaning
- TextureGAN: Controlling Deep Image Synthesis With Texture Patches (CVPR 2018 · similarity 0.88)
- GramGAN: Deep 3D Texture Synthesis From 2D Exemplars (NeurIPS 2020 · similarity 0.87)
- AUV-Net: Learning Aligned UV Maps for Texture Transfer and Synthesis (CVPR 2022 · similarity 0.87)
- TUVF: Learning Generalizable Texture UV Radiance Fields (ICLR 2024 · similarity 0.85)
- Paint-it: Text-to-Texture Synthesis via Deep Convolutional Texture Map Optimization and Physically-Based Rendering (CVPR 2024 · similarity 0.85)
- Textured Gaussians for Enhanced 3D Scene Appearance Modeling (CVPR 2025 · similarity 0.85)
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.