AutoDIR: Automatic All-in-One Image Restoration with Latent Diffusion
Abstract
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. 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: https://jiangyitong.github.io/ AutoDIR_webpage/.
Cite
Text
Jiang et al. "AutoDIR: Automatic All-in-One Image Restoration with Latent Diffusion." Proceedings of the European Conference on Computer Vision (ECCV), 2024. doi:10.1007/978-3-031-73661-2_19Markdown
[Jiang et al. "AutoDIR: Automatic All-in-One Image Restoration with Latent Diffusion." Proceedings of the European Conference on Computer Vision (ECCV), 2024.](https://mlanthology.org/eccv/2024/jiang2024eccv-autodir/) doi:10.1007/978-3-031-73661-2_19BibTeX
@inproceedings{jiang2024eccv-autodir,
title = {{AutoDIR: Automatic All-in-One Image Restoration with Latent Diffusion}},
author = {Jiang, Yitong and Zhang, Zhaoyang and Xue, Tianfan and Gu, Jinwei},
booktitle = {Proceedings of the European Conference on Computer Vision (ECCV)},
year = {2024},
doi = {10.1007/978-3-031-73661-2_19},
url = {https://mlanthology.org/eccv/2024/jiang2024eccv-autodir/}
}