CubeDiff: Repurposing Diffusion-Based Image Models for Panorama 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_ | 1866912207600615424 |
|---|---|
| author | Kalischek, Nikolai Oechsle, Michael Manhardt, Fabian Henzler, Philipp Schindler, Konrad Tombari, Federico |
| author_facet | Kalischek, Nikolai Oechsle, Michael Manhardt, Fabian Henzler, Philipp Schindler, Konrad Tombari, Federico |
| contents | We introduce a novel method for generating 360° panoramas from text prompts or images. Our approach leverages recent advances in 3D generation by employing multi-view diffusion models to jointly synthesize the six faces of a cubemap. Unlike previous methods that rely on processing equirectangular projections or autoregressive generation, our method treats each face as a standard perspective image, simplifying the generation process and enabling the use of existing multi-view diffusion models. We demonstrate that these models can be adapted to produce high-quality cubemaps without requiring correspondence-aware attention layers. Our model allows for fine-grained text control, generates high resolution panorama images and generalizes well beyond its training set, whilst achieving state-of-the-art results, both qualitatively and quantitatively. Project page: https://cubediff.github.io/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_17162 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CubeDiff: Repurposing Diffusion-Based Image Models for Panorama Generation Kalischek, Nikolai Oechsle, Michael Manhardt, Fabian Henzler, Philipp Schindler, Konrad Tombari, Federico Computer Vision and Pattern Recognition Machine Learning We introduce a novel method for generating 360° panoramas from text prompts or images. Our approach leverages recent advances in 3D generation by employing multi-view diffusion models to jointly synthesize the six faces of a cubemap. Unlike previous methods that rely on processing equirectangular projections or autoregressive generation, our method treats each face as a standard perspective image, simplifying the generation process and enabling the use of existing multi-view diffusion models. We demonstrate that these models can be adapted to produce high-quality cubemaps without requiring correspondence-aware attention layers. Our model allows for fine-grained text control, generates high resolution panorama images and generalizes well beyond its training set, whilst achieving state-of-the-art results, both qualitatively and quantitatively. Project page: https://cubediff.github.io/ |
| title | CubeDiff: Repurposing Diffusion-Based Image Models for Panorama Generation |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2501.17162 |