Research map / 3D vision / 3D generation
PBR3DGen: A VLM-Guided Mesh Generation with High-Quality PBR Texture
Xiaokang Wei, Bowen Zhang, Xianghui Yang, Yuxuan Wang, Chunchao Guo, Xi Zhao, Yan Luximon
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 VLM-guided condition (ablation) | multi-view albedo diffusion · Objaverse multi-view albedo (300 objects) | Albedo PSNR: 17.155 -> 18.186 | Improved |
| add VLM-guided condition (ablation) | multi-view albedo diffusion · Objaverse multi-view albedo (300 objects) | Albedo MSE: 0.030 -> 0.023 | Improved |
| add VLM-guided condition (ablation) | metallic-roughness diffusion · Objaverse multi-view roughness/metallic (300 objects) | Roughness PSNR: 18.023 -> 21.095 | Improved |
| add VLM-guided condition (ablation) | metallic-roughness diffusion · Objaverse multi-view roughness/metallic (300 objects) | Metallic PSNR: 15.890 -> 17.718 | Improved |
| add view-dependent illumination condition (ablation) | metallic-roughness diffusion · Objaverse multi-view roughness/metallic (300 objects) | Roughness PSNR: 19.803 -> 21.095 | Improved |
| add view-dependent illumination condition (ablation) | metallic-roughness diffusion · Objaverse multi-view roughness/metallic (300 objects) | Metallic PSNR: 15.871 -> 17.718 | Improved |
| add both VLM guide and VD-illumination (ablation) | metallic-roughness diffusion · Objaverse multi-view roughness/metallic (300 objects) | Roughness PSNR: 18.649 -> 21.095 | Improved |
| add both VLM guide and VD-illumination (ablation) | metallic-roughness diffusion · Objaverse multi-view roughness/metallic (300 objects) | Metallic PSNR: 15.293 -> 17.718 | Improved |
| use latent concatenation instead of ControlNet (ablation) | multi-view PBR diffusion · Objaverse multi-view albedo, feature injection (300 objects) | Albedo PSNR: 17.874 -> 18.186 | Improved |
| use latent concatenation instead of IP-Adapter (ablation) | multi-view PBR diffusion · Objaverse multi-view albedo, feature injection (300 objects) | Albedo PSNR: 17.351 -> 18.186 | Improved |
| encode albedo and metallic-roughness in separate latent spaces instead of five-channel input (ablation) | multi-view PBR diffusion · Objaverse multi-view albedo, feature injection (300 objects) | Albedo PSNR: 16.986 -> 18.186 | Improved |
| use PBR3DGen instead of IntrinsicAnything | IntrinsicAnything · Objaverse multi-view albedo (300 objects) | Albedo PSNR: 16.775 -> 18.186 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- SF3D — SF3D: Stable Fast 3D Mesh Reconstruction with UV-unwrapping and Illumination Disentanglement (CVPR 2025) · better on 8 of 8 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- TexGaussian: Generating High-quality PBR Material via Octree-based 3D Gaussian Splatting (CVPR 2025 · similarity 0.92)
- MeshGen: Generating PBR Textured Mesh with Render-Enhanced Auto-Encoder and Generative Data Augmentation (CVPR 2025 · similarity 0.91)
- Material Palette: Extraction of Materials from a Single Image (CVPR 2024 · similarity 0.89)
- IntrinsiX: High-Quality PBR Generation using Image Priors (NeurIPS 2025 · similarity 0.88)
- Meta 3D AssetGen: Text-to-Mesh Generation with High-Quality Geometry, Texture, and PBR Materials (NeurIPS 2024 · similarity 0.88)
- MatSynth: A Modern PBR Materials Dataset (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.
