search

LoSealL / VideoSuperResolution

A collection of state-of-the-art video or single-image super-resolution architectures, reimplemented in tensorflow.
MIT License
1.61k stars 295 forks source link

No information of mcl-v in /Data/datasets.yaml #114

Closed Susan19900316 closed 3 years ago

Susan19900316 commented 4 years ago

There is no information of mcl-v in /Data/datasets.yaml. Something is wrong when I run "python train.py vespcn --dataset mcl-v --memory_limit 1GB --epochs 100" as guided.

LoSealL commented 4 years ago

You can find it here: http://mcl.usc.edu/mcl-v-database/

kyz20 commented 3 years ago

You can find it here: http://mcl.usc.edu/mcl-v-database/

Could you please tell me how to convert the YUV file to PNG file?

LoSealL commented 3 years ago

You can find it here: http://mcl.usc.edu/mcl-v-database/

Could you please tell me how to convert the YUV file to PNG file?

@kyz20 You can refer to ffmpeg

ffmpeg -f rawvideo -s 1920x1080 -pix_fmt yuv420p -i <path-to-file.yuv> out_dir/frame_%05d.png

-s follows by the frame width x height, and -pix_fmt follows the YUV format. For this MCL-V dataset, it's yuv420p. For more information, you can refer to FOURCC