Detectron2 is FAIR's next-generation platform for object detection and segmentation.

Designing a Practical Degradation Model for Deep Blind Image Super-Resolution (PyTorch) - We released the testing code!

[CVPR 2021] Unsupervised Degradation Representation Learning for Blind Super-Resolution

[CVPR 2021] Exploring Sparsity in Image Super-Resolution for Efficient Inference

Implementation of the Swin Transformer in PyTorch.

Image Restoration Toolbox (PyTorch). Training and testing codes for USRNet, DnCNN, FFDNet, SRMD, DPSR, ESRGAN

we proposed a light-weigt but efficient blanced attention mechanism for SISR task.

Official repository for "Multi-Stage Progressive Image Restoration" (CVPR 2021). SOTA results for image deblurring, deraining, and denoising.

Code repo for "Cross-Scale Internal Graph Neural Network for Image Super-Resolution" (NeurIPS'20)

Image Enhancement Zoo for collecting deraining, dehazing and brightening methods, datasets, and codes.

ECCV18 Workshops - Enhanced SRGAN. Champion PIRM Challenge on Perceptual Super-Resolution. The training codes are in BasicSR.

Official DeiT repository

A Progressive Fusion Generative Adversarial Network for Realistic and Consistent Video Super-Resolution

GOM：New Metric for Re-identification. 👉GOM explicitly balances the effect of performing retrieval and verification into a single unified metric.

[NeurIPS'20] GradAug: A New Regularization Method for Deep Neural Networks

[Params: Only 272K!!!] Efficient Image Super-Resolution Using Pixel Attention, in ECCV Workshop, 2020.

EDSR, RCAN, SRGAN, SRFEAT, ESRGAN

categorize current rain removal methods and fairly evaluate their deraining ability based on diverse benchmark datasets

Blind Face Restoration via Deep Multi-scale Component Dictionaries (ECCV 2020)

A pytorch implementation of "Domain-Adaptive Few-Shot Learning"

This is the code for multi-exposure image fusion via generative adversarial networks