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-5

Super-Resolution

3260 papers • 126 benchmarks • 313 datasets

Super-Resolution is a task in computer vision that involves increasing the resolution of an image or video by generating missing high-frequency details from low-resolution input. The goal is to produce an output image with a higher resolution than the input image, while preserving the original content and structure. ( Credit: MemNet )

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in super-resolution-10

Trend

Dataset

Best Model

Actions

No benchmarks available.

Libraries

i

Use these libraries to find super-resolution-10 models and implementations

Datasets

MSU SR-QA Dataset

MSU SR-QA Dataset

TextZoom

RealSRSet

DRealSR

QMUL-SurvFace

Holopix50k

Subtasks

Image Super-Resolution Video Super-Resolution Multi-Frame Super-Resolution Reference-based Super-Resolution Reference-based Super-Resolution

Most implemented papers

Generative Adversarial Networks

Yoshua Bengio, Aaron C. Courville, I. Goodfellow, Mehdi Mirza, Sherjil Ozair, David Warde-Farley, Bing Xu, Jean Pouget-Abadie•Tue Dec 14 2021

A comprehensive guide to GANs, covering their architecture, loss functions, training methods, applications, evaluation metrics, challenges, and future directions is provided.

30456

Content

Introduction Benchmarks Datasets Subtasks Libraries Papers

Holopix50k

CATS

PROBA-V

Stanford Light Field

Stanford Light Field

OST300

Depth Map Super-Resolution

Reference-based Video Super-Resolution

3D Object Super-Resolution

0

Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network

Alykhan Tejani, Wenzhe Shi, Jose Caballero, Ferenc Huszár, J. Totz, Andrew P. Aitken, Zehan Wang, Lucas Theis, C. Ledig•Wed Sep 14 2016

SRGAN, a generative adversarial network (GAN) for image super-resolution (SR), is presented, to its knowledge, the first framework capable of inferring photo-realistic natural images for 4x upscaling factors and a perceptual loss function which consists of an adversarial loss and a content loss.

11742 0

Perceptual Losses for Real-Time Style Transfer and Super-Resolution

Justin Johnson, Li Fei-Fei, Alexandre Alahi•Sat Mar 26 2016

This work considers image transformation problems, and proposes the use of perceptual loss functions for training feed-forward networks for image transformation tasks, and shows results on image style transfer, where aFeed-forward network is trained to solve the optimization problem proposed by Gatys et al. in real-time.

11126 0

Image Super-Resolution Using Deep Convolutional Networks

Chen Change Loy, Xiaoou Tang, Kaiming He, Chao Dong•Tue Dec 30 2014

This work proposes a deep learning method for single image super-resolution (SR) that directly learns an end-to-end mapping between the low/high-resolution images, represented as a deep convolutional neural network (CNN) that takes the low-resolution image as the input and outputs the high-resolution one.

8773 0

Enhanced Deep Residual Networks for Single Image Super-Resolution

Bee Lim, Sanghyun Son, Heewon Kim, Seungjun Nah, Kyoung Mu Lee•Sun Jul 09 2017

This paper develops an enhanced deep super-resolution network (EDSR) with performance exceeding those of current state-of-the-art SR methods, and proposes a new multi-scale deepsuper-resolution system (MDSR) and training method, which can reconstruct high-resolution images of different upscaling factors in a single model.

6646 0

ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks

Y. Qiao, Chen Change Loy, Xiaoou Tang, Chao Dong, Xintao Wang, Jinjin Gu, Yihao Liu, Ke Yu, Shixiang Wu•Fri Aug 31 2018

This work thoroughly study three key components of SRGAN – network architecture, adversarial loss and perceptual loss, and improves each of them to derive an Enhanced SRGAN (ESRGAN), which achieves consistently better visual quality with more realistic and natural textures than SRGAN.

4394 0

SinGAN: Learning a Generative Model From a Single Natural Image

Tali Dekel, T. Michaeli, Tamar Rott Shaham•Wed May 01 2019

SinGAN, an unconditional generative model that can be learned from a single natural image, is introduced, trained to capture the internal distribution of patches within the image, and is then able to generate high quality, diverse samples that carry the same visual content as the image.

913 0

Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network

D. Rueckert, Wenzhe Shi, Jose Caballero, Ferenc Huszár, J. Totz, Andrew P. Aitken, Rob Bishop, Zehan Wang•Sun Jun 26 2016

This paper presents the first convolutional neural network capable of real-time SR of 1080p videos on a single K2 GPU and introduces an efficient sub-pixel convolution layer which learns an array of upscaling filters to upscale the final LR feature maps into the HR output.

5749 0

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 0

High-Resolution Image Synthesis with Latent Diffusion Models

A. Blattmann, Robin Rombach, Dominik Lorenz, B. Ommer, Patrick Esser•Sun Dec 19 2021

These latent diffusion models achieve new state of the art scores for image inpainting and class-conditional image synthesis and highly competitive performance on various tasks, including unconditional image generation, text-to-image synthesis, and super-resolution, while significantly reducing computational requirements compared to pixel-based DMs.

21712 0

Adding a benchmark result helps the community track progress.

Super-Resolution | State-of-the-Art