6c6e82406e
The intuition is this will help guide the network to make better informed decisions about how it performs upsampling based on how it perceives the underlying content. (I'm giving up on letting networks detect their own quality - I'm not convinced it is actually feasible) |
||
---|---|---|
.. | ||
byol | ||
classifiers | ||
diffusion | ||
fixup_resnet | ||
flownet2@db2b7899ea | ||
glean | ||
global_convs | ||
optical_flow | ||
pixel_level_contrastive_learning | ||
segformer | ||
srflow | ||
styled_sr | ||
stylegan | ||
switched_conv | ||
tecogan | ||
transformers | ||
vqvae | ||
__init__.py | ||
arch_util.py | ||
discriminator_vgg_arch.py | ||
feature_arch.py | ||
lightweight_gan.py | ||
ResGen_arch.py | ||
RRDBNet_arch.py | ||
spinenet_arch.py |