PBR-Inspired Controllable Diffusion for Image Generation
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866908818673238016 |
|---|---|
| author | Xue, Bowen Guarnera, Giuseppe Claudio Zhao, Shuang Montazeri, Zahra |
| author_facet | Xue, Bowen Guarnera, Giuseppe Claudio Zhao, Shuang Montazeri, Zahra |
| contents | Despite recent advances in text-to-image generation, controlling geometric layout and PBR material properties in synthesized scenes remains challenging. We present a pipeline that first produces a G-buffer (albedo, normals, depth, roughness, shading, and metallic) from a text prompt and then renders a final image through a PBR-inspired branch network. This intermediate representation enables fine-grained control: users can copy and paste within specific G-buffer channels to insert or reposition objects, or apply masks to the irradiance channel to adjust lighting locally. As a result, real objects can be seamlessly integrated into virtual scenes. By separating user-friendly scene description from image rendering, our method offers a practical balance between detailed post-generation control and efficient text-driven synthesis. We demonstrate its effectiveness through quantitative evaluations and a user study with 156 participants, showing consistent human preference over strong baselines and confirming that G-buffer control extends the flexibility of text-guided image generation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_15147 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | PBR-Inspired Controllable Diffusion for Image Generation Xue, Bowen Guarnera, Giuseppe Claudio Zhao, Shuang Montazeri, Zahra Graphics Despite recent advances in text-to-image generation, controlling geometric layout and PBR material properties in synthesized scenes remains challenging. We present a pipeline that first produces a G-buffer (albedo, normals, depth, roughness, shading, and metallic) from a text prompt and then renders a final image through a PBR-inspired branch network. This intermediate representation enables fine-grained control: users can copy and paste within specific G-buffer channels to insert or reposition objects, or apply masks to the irradiance channel to adjust lighting locally. As a result, real objects can be seamlessly integrated into virtual scenes. By separating user-friendly scene description from image rendering, our method offers a practical balance between detailed post-generation control and efficient text-driven synthesis. We demonstrate its effectiveness through quantitative evaluations and a user study with 156 participants, showing consistent human preference over strong baselines and confirming that G-buffer control extends the flexibility of text-guided image generation. |
| title | PBR-Inspired Controllable Diffusion for Image Generation |
| topic | Graphics |
| url | https://arxiv.org/abs/2503.15147 |