REDS Realistic and Dynamic Scenes Dataset

REDS stands for The REalistic and Diverse Scenes. It is a dataset that provides realistic and dynamic scenes and can be used for video deblurring and super-resolution.

The dataset consists of 300 video sequences with a resolution of 720 × 1,280, each video has 100 frames, of which the training set, validation set and test set have 240, 30 and 30 videos respectively.