SPAN: Swift Parameter-free Attention Network for Efficient Super-Resolution

Mirror for all official pretrained SPAN models. This includes the 2x and 4x versions of ch48 and ch52 models.

https://github.com/hongyuanyu/span

RGT: Recursive Generalization Transformer for Image Super-Resolution

Mirror for all official pretrained RGT models. This includes the 2x, 3x, and 4x versions of RGT and RGT-S.

https://github.com/zhengchen1999/RGT

OmniSR: Omni Aggregation Networks for Lightweight Image Super-Resolution

Mirror for all official pretrained OmniSR models. This includes the 2x, 3x, and 4x versions trained on DF2K and DIV2K.

https://github.com/Francis0625/Omni-SR

KBNet: Kernel Basis Network for Image Restoration

Mirror for all official pretrained KBNet models.

https://github.com/zhangyi-3/KBNet

HAT: Activating More Pixels in Image Super-Resolution Transformer

Mirror for all official pretrained HAT models. This includes the 2x, 3x, and 4x versions of HAT, HAT-S, and HAT-L.

https://github.com/XPixelGroup/HAT

DAT: Dual Aggregation Transformer for Image Super-Resolution

Mirror for all official pretrained DAT models. This includes the 2x, 3x, and 4x versions of DAT, DAT-S, DAT light, and DAT-2.

https://github.com/zhengchen1999/DAT