Dataset | Year | Type | Download Link | Video Number | Resolution | Color Space |
V-train1 | – | Train | https://media.xiph.org/video/derf/ | 25 | – | YUV |
V-test | – | Test | https://media.xiph.org/video/derf/ | 5 | – | YUV |
Vid4 | 2011 | Test | https://drive.google.com/drive/folders/10-gUO6zBeOpWE amrWKCtSkkUFukB9W5m | 4 | 720 × 480 | RGB |
YUV21 | 2014 | Test | http://www.codersvoice.com/a/webbase/video/08/152014/ 130.html | 21 | 352 × 288 | YUV |
Venice | 2014 | Train | https://www.harmonicinc.com/free-4k-demo-footage/ | 1 | 3, 840 × 2, 160 | RGB |
Myanmar | 2014 | Train | https://www.harmonicinc.com/insights/blog/4k-in-context / | 1 | 3, 840 × 2, 160 | RGB |
CDVL | 2016 | Train | http://www.cdvl.org/ | – | 1, 920 × 1, 080 | RGB |
Vimeo90K | 2019 | Train+Test | http://toflow.csail.mit.edu/ | 91,701 | 448 × 256 | RGB |
REDS | 2019 | Train+Test | https://seungjunnah.github.io/Datasets/reds.html | 270 | 1, 920 × 1, 080 | RGB |
UVG | 2020 | Test | http://ultravideo.cs.tut.fi/ | 16 | 3, 840 × 2, 160 | YUV |
Vid4
Videoset4 consists of the four test videos: walk, foliage, city, and calendar.
Vimeo90k
This dataset consists of 89,800 video clips downloaded from vimeo.com, which covers large variaty of scenes and actions. It is designed for the following four video processing tasks: temporal frame interpolation, video denoising, video deblocking, and video super-resolution.
UVG
The open Ultra Video Group (UVG) dataset is composed of 16 versatile 4K (3840×2160) test video sequences captured at 50/120 fps.
Challenges
Name | Year | Organizer | Location | Website | Dataset | Scale | Champion | PSNR | SSIM |
NTIRE 2019 Video Restoration and Enhancement Challenges | 2019 | CVPR | Long Beach, California | https://data.vision.ee.e thz.ch/cvl/ntire19/ | REDS | ×4 | EDVR [24] | 31.79 | 0.8962 |
YOUKU Video Super-Resolution and Enhancement Challenge | 2019 | Alibaba | Hangzhou, China | https://tianchi.aliyun.c om/competition/entra nce/231711/introductio n | Youku-VESR | ×4 | VESR-Net [84] | 37.85 | – |
AIM 2019 Challenge on Video Extreme Super-Resolution | 2019 | ECCV | Hong Kong, China | https://www.aim2019. org/ | Vid3oC | ×16 | EDVR+ | 22.53 | 0.64 |
Mobile Restoration ChallengeVideo | 2019 | ICIP & Kwai | – | https://www.kuaishou .com/activity/icip2019 | – | – | – | – | – |
AIM 2020 Challenge on Video Extreme Super-Resolution | 2020 | ECCV | Boston, Massachusetts | http://aim2020.org/ | – | ×16 | – | – | – |