Home Research Papers Datasets State of the Art Pricing

Discover, visualize, and connect AI research papers. Explore the latest trends and insights in artificial intelligence research.

Product

Home
Research Papers
About

Support

Contact
Terms of Service
Privacy Policy

© 2026 Papersgraph. All rights reserved.

computer-vision-10

JPEG Artifact Correction

3260 papers • 126 benchmarks • 313 datasets

Correction of visual artifacts caused by JPEG compression, these artifacts are usually grouped into three types: blocking, blurring, and ringing. They are caused by quantization and removal of high frequency DCT coefficients.

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in jpeg-artifact-correction-10

Trend

Dataset

Best Model

Actions

Live1 (Quality 10 Grayscale)

Live1 (Quality 10 Grayscale)

LIVE1 (Quality 20 Grayscale)

LIVE1 (Quality 20 Grayscale)

Libraries

i

Use these libraries to find jpeg-artifact-correction-10 models and implementations

Datasets

DIV2K

BSDS500

LIVE (Public-Domain Subjective Image Quality Database)

LIVE (Public-Domain Subjective Image Quality Database)

Classic5

ICB

Subtasks

No subtasks available.

Most implemented papers

Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising

Deyu Meng, K. Zhang, W. Zuo, Yunjin Chen, Lei Zhang•Fri Aug 12 2016

This paper investigates the construction of feed-forward denoising convolutional neural networks (DnCNNs) to embrace the progress in very deep architecture, learning algorithm, and regularization method into image Denoising, and uses residual learning and batch normalization to speed up the training process as well as boost theDenoising performance.

7728

Content

Introduction Benchmarks Datasets Subtasks Libraries Papers

LIVE1 (Quality 10 Color)

LIVE1 (Quality 10 Color)

LIVE1 (Quality 20 Color)

LIVE1 (Quality 20 Color)

LIVE1 (Quality 30 Grayscale)

LIVE1 (Quality 30 Grayscale)

Classic5 (Quality 10 Grayscale)

Classic5 (Quality 10 Grayscale)

ICB (Quality 10 Color)

ICB (Quality 10 Color)

ICB (Quality 20 Color)

ICB (Quality 20 Color)

Classic5 (Quality 20 Grayscale)

Classic5 (Quality 20 Grayscale)

Classic5 (Quality 30 Grayscale)

Classic5 (Quality 30 Grayscale)

ICB (Quality 20 Grayscale)

ICB (Quality 20 Grayscale)

ICB (Quality 10 Grayscale)

ICB (Quality 10 Grayscale)

LIVE1 (Quality 40 Grayscale)

LIVE1 (Quality 40 Grayscale)

Classic5 (Quality 40 Grayscale)

Classic5 (Quality 40 Grayscale)

ICB (Quality 30 Color)

ICB (Quality 30 Color)

LIVE1 (Quality 30 Color)

LIVE1 (Quality 30 Color)

BSDS500 (Quality 10 Color)

BSDS500 (Quality 10 Color)

BSDS500 (Quality 20 Color)

BSDS500 (Quality 20 Color)

BSDS500 (Quality 30 Color)

BSDS500 (Quality 30 Color)

BSDS500 (Quality 10 Grayscale)

BSDS500 (Quality 10 Grayscale)

BSDS500 (Quality 20 Grayscale)

BSDS500 (Quality 20 Grayscale)

BSDS500 (Quality 30 Grayscale)

BSDS500 (Quality 30 Grayscale)

ICB (Quality 30 Grayscale)

ICB (Quality 30 Grayscale)

Live1 (Quality 50 Grayscale)

Live1 (Quality 50 Grayscale)

Live1 (Quality 70 Grayscale)

Live1 (Quality 70 Grayscale)

Live1 (Quality 80 Grayscale)

Live1 (Quality 80 Grayscale)

Urban100 (Quality 10 Grayscale)

Urban100 (Quality 10 Grayscale)

Urban100 (Quality 20 Grayscale)

Urban100 (Quality 20 Grayscale)

Urban100 (Quality 30 Grayscale)

Urban100 (Quality 30 Grayscale)

Urban100 (Quality 40 Grayscale)

Urban100 (Quality 40 Grayscale)

LIVE1 (Quality 40 Color)

LIVE1 (Quality 40 Color)

BSD500 (Quality 10 Color)

BSD500 (Quality 10 Color)

BSD500 (Quality 20 Color)

BSD500 (Quality 20 Color)

BSD500 (Quality 30 Color)

BSD500 (Quality 30 Color)

BSD500 (Quality 40 Color)

BSD500 (Quality 40 Color)

Urban100 (Quality 10 Color)

Urban100 (Quality 10 Color)

Urban100 (Quality 20 Color)

Urban100 (Quality 20 Color)

Urban100 (Quality 30 Color)

Urban100 (Quality 30 Color)

Urban100 (Quality 40 Color)

Urban100 (Quality 40 Color)

0

Multi-level Wavelet-CNN for Image Restoration

K. Zhang, W. Zuo, Liang Lin, Hongzhi Zhang, Pengju Liu•Thu May 17 2018

This paper presents a novel multi-level wavelet CNN model for better tradeoff between receptive field size and computational efficiency, and shows the effectiveness of MWCNN for image denoising, single image super-resolution, and JPEG image artifacts removal.

746 0

Image Restoration Using Convolutional Auto-encoders with Symmetric Skip Connections

Chunhua Shen, Xiao-Jiao Mao, Yubin Yang•Tue Jun 28 2016

To deal with the problem that deeper networks tend to be more difficult to train, this work proposes to symmetrically link convolutional and deconvolutional layers with skip-layer connections, with which the training converges much faster and attains better results.

425 0

Swin2SR: SwinV2 Transformer for Compressed Image Super-Resolution and Restoration

Radu Timofte, Marcos V. Conde, Ui-Jin Choi, Maxime Burchi•Wed Sep 21 2022

The novel Swin Transformer V2 is explored, to improve SwinIR for image super-resolution, and in particular, the compressed input scenario, and the Swin2SR method, which is a top-5 solution at the AIM 2022 Challenge on Super-Resolution of Compressed Image and Video.

197 0

Compression Artifacts Reduction by a Deep Convolutional Network

Chen Change Loy, Xiaoou Tang, Chao Dong, Yubin Deng•Sun Apr 26 2015

A compact and efficient network for seamless attenuation of different compression artifacts is formulated and it is demonstrated that a deeper model can be effectively trained with the features learned in a shallow network.

846 0

Residual Dense Network for Image Restoration

Yapeng Tian, Yu Kong, Y. Fu, Yulun Zhang, Bineng Zhong•Mon Dec 24 2018

This work proposes residual dense block (RDB) to extract abundant local features via densely connected convolutional layers and proposes local feature fusion in RDB to adaptively learn more effective features from preceding and current local features and stabilize the training of wider network.

819 0

MemNet: A Persistent Memory Network for Image Restoration

Ying Tai, Jian Yang, Xiaoming Liu, Chunyan Xu•Sun Aug 06 2017

A very deep persistent memory network (MemNet) is proposed that introduces a memory block, consisting of a recursive unit and a gate unit, to explicitly mine persistent memory through an adaptive learning process.

1706 0

Quantization Guided JPEG Artifact Correction

Abhinav Shrivastava, Ser-Nam Lim, L. Davis, Max Ehrlich•Thu Apr 16 2020

This work creates a novel architecture which is parameterized by the JPEG files quantization matrix, which allows a single model to achieve state-of-the-art performance over models trained for specific quality settings.

118 0

Towards Flexible Blind JPEG Artifacts Removal

K. Zhang, Radu Timofte, Jiaxi Jiang•Tue Sep 28 2021

This paper proposes a flexible blind convolutional neural network that can predict the adjustable quality factor to control the trade-off between artifacts removal and details preservation and achieves favorable performance against state-of-the-art methods in terms of both quantitative metrics and visual quality.

136 0

Learning a Single Model With a Wide Range of Quality Factors for JPEG Image Artifacts Removal

Yongtao Wang, Jianwei Li, Haihua Xie, K. Ma•Thu Sep 03 2020

The proposed network is a single model approach that can be trained for handling a wide range of quality factors while consistently delivering superior or comparable image artifacts removal performance.

46 0

Adding a benchmark result helps the community track progress.

JPEG Artifact Correction | State-of-the-Art