IntrinsicEdit: Precise generative image manipulation in intrinsic space
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909610809491456 |
|---|---|
| author | Lyu, Linjie Deschaintre, Valentin Hold-Geoffroy, Yannick Hašan, Miloš Yoon, Jae Shin Leimkühler, Thomas Theobalt, Christian Georgiev, Iliyan |
| author_facet | Lyu, Linjie Deschaintre, Valentin Hold-Geoffroy, Yannick Hašan, Miloš Yoon, Jae Shin Leimkühler, Thomas Theobalt, Christian Georgiev, Iliyan |
| contents | Generative diffusion models have advanced image editing with high-quality results and intuitive interfaces such as prompts and semantic drawing. However, these interfaces lack precise control, and the associated methods typically specialize on a single editing task. We introduce a versatile, generative workflow that operates in an intrinsic-image latent space, enabling semantic, local manipulation with pixel precision for a range of editing operations. Building atop the RGB-X diffusion framework, we address key challenges of identity preservation and intrinsic-channel entanglement. By incorporating exact diffusion inversion and disentangled channel manipulation, we enable precise, efficient editing with automatic resolution of global illumination effects -- all without additional data collection or model fine-tuning. We demonstrate state-of-the-art performance across a variety of tasks on complex images, including color and texture adjustments, object insertion and removal, global relighting, and their combinations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_08889 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | IntrinsicEdit: Precise generative image manipulation in intrinsic space Lyu, Linjie Deschaintre, Valentin Hold-Geoffroy, Yannick Hašan, Miloš Yoon, Jae Shin Leimkühler, Thomas Theobalt, Christian Georgiev, Iliyan Graphics Computer Vision and Pattern Recognition Generative diffusion models have advanced image editing with high-quality results and intuitive interfaces such as prompts and semantic drawing. However, these interfaces lack precise control, and the associated methods typically specialize on a single editing task. We introduce a versatile, generative workflow that operates in an intrinsic-image latent space, enabling semantic, local manipulation with pixel precision for a range of editing operations. Building atop the RGB-X diffusion framework, we address key challenges of identity preservation and intrinsic-channel entanglement. By incorporating exact diffusion inversion and disentangled channel manipulation, we enable precise, efficient editing with automatic resolution of global illumination effects -- all without additional data collection or model fine-tuning. We demonstrate state-of-the-art performance across a variety of tasks on complex images, including color and texture adjustments, object insertion and removal, global relighting, and their combinations. |
| title | IntrinsicEdit: Precise generative image manipulation in intrinsic space |
| topic | Graphics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.08889 |