RGB$\leftrightarrow$X: Image decomposition and synthesis using material- and lighting-aware diffusion models

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Zeng, Zheng, Deschaintre, Valentin, Georgiev, Iliyan, Hold-Geoffroy, Yannick, Hu, Yiwei, Luan, Fujun, Yan, Ling-Qi, Hašan, Miloš
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913338322059264
author Zeng, Zheng
Deschaintre, Valentin
Georgiev, Iliyan
Hold-Geoffroy, Yannick
Hu, Yiwei
Luan, Fujun
Yan, Ling-Qi
Hašan, Miloš
author_facet Zeng, Zheng
Deschaintre, Valentin
Georgiev, Iliyan
Hold-Geoffroy, Yannick
Hu, Yiwei
Luan, Fujun
Yan, Ling-Qi
Hašan, Miloš
contents The three areas of realistic forward rendering, per-pixel inverse rendering, and generative image synthesis may seem like separate and unrelated sub-fields of graphics and vision. However, recent work has demonstrated improved estimation of per-pixel intrinsic channels (albedo, roughness, metallicity) based on a diffusion architecture; we call this the RGB$\rightarrow$X problem. We further show that the reverse problem of synthesizing realistic images given intrinsic channels, X$\rightarrow$RGB, can also be addressed in a diffusion framework. Focusing on the image domain of interior scenes, we introduce an improved diffusion model for RGB$\rightarrow$X, which also estimates lighting, as well as the first diffusion X$\rightarrow$RGB model capable of synthesizing realistic images from (full or partial) intrinsic channels. Our X$\rightarrow$RGB model explores a middle ground between traditional rendering and generative models: we can specify only certain appearance properties that should be followed, and give freedom to the model to hallucinate a plausible version of the rest. This flexibility makes it possible to use a mix of heterogeneous training datasets, which differ in the available channels. We use multiple existing datasets and extend them with our own synthetic and real data, resulting in a model capable of extracting scene properties better than previous work and of generating highly realistic images of interior scenes.
format Preprint
id arxiv_https___arxiv_org_abs_2405_00666
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle RGB$\leftrightarrow$X: Image decomposition and synthesis using material- and lighting-aware diffusion models
Zeng, Zheng
Deschaintre, Valentin
Georgiev, Iliyan
Hold-Geoffroy, Yannick
Hu, Yiwei
Luan, Fujun
Yan, Ling-Qi
Hašan, Miloš
Computer Vision and Pattern Recognition
Graphics
The three areas of realistic forward rendering, per-pixel inverse rendering, and generative image synthesis may seem like separate and unrelated sub-fields of graphics and vision. However, recent work has demonstrated improved estimation of per-pixel intrinsic channels (albedo, roughness, metallicity) based on a diffusion architecture; we call this the RGB$\rightarrow$X problem. We further show that the reverse problem of synthesizing realistic images given intrinsic channels, X$\rightarrow$RGB, can also be addressed in a diffusion framework. Focusing on the image domain of interior scenes, we introduce an improved diffusion model for RGB$\rightarrow$X, which also estimates lighting, as well as the first diffusion X$\rightarrow$RGB model capable of synthesizing realistic images from (full or partial) intrinsic channels. Our X$\rightarrow$RGB model explores a middle ground between traditional rendering and generative models: we can specify only certain appearance properties that should be followed, and give freedom to the model to hallucinate a plausible version of the rest. This flexibility makes it possible to use a mix of heterogeneous training datasets, which differ in the available channels. We use multiple existing datasets and extend them with our own synthetic and real data, resulting in a model capable of extracting scene properties better than previous work and of generating highly realistic images of interior scenes.
title RGB$\leftrightarrow$X: Image decomposition and synthesis using material- and lighting-aware diffusion models
topic Computer Vision and Pattern Recognition
Graphics
url https://arxiv.org/abs/2405.00666