REalistic and Diverse Scenes dataset realistic and dynamic scenes
The realistic and dynamic scenes (REDS) dataset was proposed in the NTIRE19 Challenge. The dataset is composed of 300 video sequences with resolution of 720×1,280, and each video has 100 frames, where the training set, the validation set and the testing set have 240, 30 and 30 videos, respectively
Source: Video Super Resolution Based on Deep Learning: A comprehensive survey
Image Source: https://seungjunnah.github.io/Datasets/reds.html
Variants: REDS
This dataset is used in 1 benchmark:
| Task | Model | Paper | Date |
|---|---|---|---|
| Deblurring | VRT | VRT: A Video Restoration Transformer | 2022-01-28 |
| Deblurring | EDVR_Deblur | EDVR: Video Restoration with Enhanced … | 2019-05-07 |
| Deblurring | DeblurGAN | DeblurGAN: Blind Motion Deblurring Using … | 2017-11-19 |
Recent papers with results on this dataset: