computer-vision-5

Unified Image Restoration

3260 papers • 126 benchmarks • 313 datasets

Using a single model to restore inputs with different degradation types.

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in unified-image-restoration-5

Trend

Dataset

Best Model

Actions

Rain100L

RESIDE

GoPro

Libraries

i

Use these libraries to find unified-image-restoration-5 models and implementations

Datasets

BSD

GoPro

LOL

Synthetic Rain Datasets

RESIDE

AWMM-100k

Subtasks

Blind All-in-One Image Restoration

Most implemented papers

Photo-Realistic Image Restoration in the Wild with Controlled Vision-Language Models

Fredrik K. Gustafsson, Zheng Zhao, Ziwei Luo, Jens Sjölund, Thomas B. Schön•Sun Apr 14 2024

A capable vision-language model and a synthetic degradation pipeline are leveraged to learn image restoration in the wild (wild IR) and a posterior sampling strategy is presented for fast noise-free image generation for various degradations.

25

Content

LOL

BSD68 sigma25

0

Paper Graph

Degradation-Aware Residual-Conditioned Optimal Transport for Unified Image Restoration

Xiaole Tang, Xiang Gu, Xiaoyi He, Xin Hu, Jian Sun•Sat Nov 02 2024

A Degradation-Aware Residual-Conditioned Optimal Transport (DA-RCOT) approach that models (all-in-one) image restoration as an optimal transport (OT) problem for unpaired and paired settings, introducing the transport residual as a degradation-specific cue for both the transport cost and the transport map.

10 0

Paper Graph

ProRes: Exploring Degradation-aware Visual Prompt for Universal Image Restoration

Tianheng Cheng, Xinggang Wang, L. Zhang, Qian Zhang, Jiaqi Ma, Guoli Wang•Thu Jun 22 2023

Dgradation-aware Visual Prompts are presented, which encode various types of image degradation, e.g., noise and blur, into unified visual prompts, which provide control over image processing and allow weighted combinations for customized image restoration.

73 0

Paper Graph

Controlling Vision-Language Models for Multi-Task Image Restoration

Ziwei Luo, Fredrik K. Gustafsson, Jens Sjolund, Zheng Zhao, Thomas B. Schon•Sun Oct 01 2023

This paper presents a degradation-aware vision-language model (DA-CLIP) to better transfer pretrained vision-language models to low-level vision tasks as a multi-task framework for image restoration.

83 0

Paper Graph

Neural Degradation Representation Learning for All-in-One Image Restoration

Mingde Yao, Ruikang Xu, Zhiwei Xiong, Yuanshen Guan, Jie Huang•Wed Oct 18 2023

This paper proposes to learn a neural degradation representation (NDR) that captures the underlying characteristics of various degradations that adaptively decomposes different types of degradations, similar to a neural dictionary that represents basic degradation components.

60 0

Paper Graph

CRNet: A Detail-Preserving Network for Unified Image Restoration and Enhancement Task

Qingsen Yan, Kangzhen Yang, Tao Hu, Genggeng Chen, Kexin Dai, Peng Wu, Yanning Zhang, Yu Cao, Wei Dong•Sun Apr 21 2024

The Composite Refinement Network (CRNet) is proposed, which can perform unified image restoration and enhancement on multiple exposure images using multiple exposure images, and explicitly separates and strengthens high and low-frequency information through pooling layers, using specially designed Multi-Branch Blocks.

17 0

Paper Graph

UIR-LoRA: Achieving Universal Image Restoration through Multiple Low-Rank Adaptation

Cheng Zhang, Jinqiu Sun, Dong Gong, Jiumei He, Yu Zhu, Yanning Zhang•Sun Sep 29 2024

This work proposes a universal image restoration framework based on multiple low-rank adapters (LoRA) from multi-domain transfer learning that leverages the pre-trained generative model as the shared component for multi-degradation restoration and transfers it to specific degradation image restoration tasks using low-rank adaptation.

1 0

Paper Graph

Unified Image Restoration and Enhancement: Degradation Calibrated Cycle Reconstruction Diffusion Model

Minglong Xue, Jinhong He, P. Shivakumara, Mingliang Zhou•Wed Dec 18 2024

CycleRDM is a novel framework designed to unify restoration and enhancement tasks while achieving high-quality mapping, and can be effectively generalized to a wide range of image restoration and enhancement tasks while requiring only a small number of training samples to be significantly superior on various benchmarks.

5 0

Paper Graph

Beyond Degradation Redundancy: Contrastive Prompt Learning for All-in-One Image Restoration

Gang Wu, Junjun Jiang, Kui Jiang, Xianming Liu, Liqiang Nie•Sun Apr 13 2025

This work introduces Contrastive Prompt Learning (CPL), a novel framework that fundamentally enhances prompt-task alignment through two complementary innovations: a Sparse Prompt Module (SPM) that efficiently captures degradation-specific features while minimizing redundancy, and a Contrastive Prompt Regularization (CPR) that explicitly strengthens task boundaries by incorporating negative prompt samples across different degradation types.

3 0

Paper Graph

Boosting All-in-One Image Restoration via Self-Improved Privilege Learning

Gang Wu, Junjun Jiang, Kui Jiang, Xianming Liu•Thu May 29 2025

Self-Improved Privilege Learning is introduced, a novel paradigm that overcomes limitations by innovatively extending the utility of privileged information beyond training into the inference stage and can be seamlessly integrated into various backbone architectures, offering substantial performance improvements with minimal computational overhead.

2 0

Paper Graph

Adding a benchmark result helps the community track progress.