search

The-Learning-And-Vision-Atelier-LAVA / SOF-VSR

[ACCV 2018 & TIP 2020] Deep Video Super-resolution using HR Optical Flow Estimation
169 stars 43 forks source link

Availability of training data? #9

Closed mwalczyk closed 5 years ago

mwalczyk commented 5 years ago

Hi,

Is the data that was used for training hosted anywhere?

I understand that the data was taken from https://www.cdvl.org/, but which videos were used?

LongguangWang commented 5 years ago

Sorry for the inconvenience. Due to the copyright issue, we cannot release the training data. We search for videos with 1080p resolution and download them except repeated ones. Then we manually extract 145 video clips from them.