A collection of state-of-the-art video or single-image super-resolution architectures, reimplemented in tensorflow.
Major change:
- Add more options to control training, for more details, please check by --helpfull
.
- Add --ensemble mode when doing inference, to learn more about "ensemble", see EDSR paper.
Bug fix:
-f
, -f2
, -f3
to add custom callbacks together