AutoDIR: Automatic All-in-One Image Restoration with Latent Diffusion

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Jiang, Yitong, Zhang, Zhaoyang, Xue, Tianfan, Gu, Jinwei
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911890648596480
author Jiang, Yitong
Zhang, Zhaoyang
Xue, Tianfan
Gu, Jinwei
author_facet Jiang, Yitong
Zhang, Zhaoyang
Xue, Tianfan
Gu, Jinwei
contents We present AutoDIR, an innovative all-in-one image restoration system incorporating latent diffusion. AutoDIR excels in its ability to automatically identify and restore images suffering from a range of unknown degradations. AutoDIR offers intuitive open-vocabulary image editing, empowering users to customize and enhance images according to their preferences. Specifically, AutoDIR consists of two key stages: a Blind Image Quality Assessment (BIQA) stage based on a semantic-agnostic vision-language model which automatically detects unknown image degradations for input images, an All-in-One Image Restoration (AIR) stage utilizes structural-corrected latent diffusion which handles multiple types of image degradations. Extensive experimental evaluation demonstrates that AutoDIR outperforms state-of-the-art approaches for a wider range of image restoration tasks. The design of AutoDIR also enables flexible user control (via text prompt) and generalization to new tasks as a foundation model of image restoration. Project is available at: \url{https://jiangyitong.github.io/AutoDIR_webpage/}.
format Preprint
id arxiv_https___arxiv_org_abs_2310_10123
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle AutoDIR: Automatic All-in-One Image Restoration with Latent Diffusion
Jiang, Yitong
Zhang, Zhaoyang
Xue, Tianfan
Gu, Jinwei
Computer Vision and Pattern Recognition
We present AutoDIR, an innovative all-in-one image restoration system incorporating latent diffusion. AutoDIR excels in its ability to automatically identify and restore images suffering from a range of unknown degradations. AutoDIR offers intuitive open-vocabulary image editing, empowering users to customize and enhance images according to their preferences. Specifically, AutoDIR consists of two key stages: a Blind Image Quality Assessment (BIQA) stage based on a semantic-agnostic vision-language model which automatically detects unknown image degradations for input images, an All-in-One Image Restoration (AIR) stage utilizes structural-corrected latent diffusion which handles multiple types of image degradations. Extensive experimental evaluation demonstrates that AutoDIR outperforms state-of-the-art approaches for a wider range of image restoration tasks. The design of AutoDIR also enables flexible user control (via text prompt) and generalization to new tasks as a foundation model of image restoration. Project is available at: \url{https://jiangyitong.github.io/AutoDIR_webpage/}.
title AutoDIR: Automatic All-in-One Image Restoration with Latent Diffusion
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2310.10123