search

ryanxingql / stdf-pytorch

Implementation of "Spatio-Temporal Deformable Convolution for Compressed Video Quality Enhancement" (AAAI'20).
https://www.aiide.org/ojs/index.php/AAAI/article/view/6697
Apache License 2.0
155 stars 20 forks source link

about vimeo 90k dataset #23

Closed doudoublues closed 1 year ago

doudoublues commented 1 year ago

why use all-intra mode to compress data? In this case, all frames have same quality, is it useful to use neighbor frames' information?

ryanxingql commented 1 year ago

Yes it is better to use LDP mode. It is my mistake. The MFQEv2 dataset is correctly compressed by LDP.

doudoublues commented 1 year ago

can i just use the same process pipeline as MFQEv2 dataset to compressed the vimeo90k data?

ryanxingql commented 1 year ago

Yes for sure. Note that you may first have to convert the Vimeo90K images into sequences (videos).

doudoublues commented 1 year ago

thx very much!