Image Super Resolution On Urban100 3X

Metrics

PSNR

SSIM

Results

Performance results of various models on this benchmark

Model Name	PSNR	SSIM	Paper Title	Repository
SwinFIR	30.43	0.8913	SwinFIR: Revisiting the SwinIR with Fast Fourier Convolution and Improved Training for Image Super-Resolution
HMA†	31.00	0.8984	HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution
HAT	30.70	0.8949	Activating More Pixels in Image Super-Resolution Transformer
Hi-IR-L	31.07	0.902	Hierarchical Information Flow for Generalized Efficient Image Restoration	-
IMDN	28.17	-	Lightweight Image Super-Resolution with Information Multi-distillation Network
DnCNN-3	27.15	-	Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
CPAT+	30.63	0.8934	Channel-Partitioned Windowed Attention And Frequency Learning for Single Image Super-Resolution	-
ML-CrAIST-Li	28.73	0.8651	ML-CrAIST: Multi-scale Low-high Frequency Information-based Cross black Attention with Image Super-resolving Transformer
HAN+	29.21	0.8710	Single Image Super-Resolution via a Holistic Attention Network
SRFBN	28.73	-	Feedback Network for Image Super-Resolution
FACD	28.818	-	Feature-domain Adaptive Contrastive Distillation for Efficient Single Image Super-Resolution	-
IPT	29.49	-	Pre-Trained Image Processing Transformer
CSNLN	29.13	0.8712	Image Super-Resolution with Cross-Scale Non-Local Attention and Exhaustive Self-Exemplars Mining
LCSCNet	27.24	-	LCSCNet: Linear Compressing Based Skip-Connecting Network for Image Super-Resolution
LTE	29.41	-	Local Texture Estimator for Implicit Representation Function
HAT-L	30.92	0.8981	Activating More Pixels in Image Super-Resolution Transformer
ML-CrAIST	28.89	0.8676	ML-CrAIST: Multi-scale Low-high Frequency Information-based Cross black Attention with Image Super-resolving Transformer
SwinOIR	28.87	0.8674	Resolution Enhancement Processing on Low Quality Images Using Swin Transformer Based on Interval Dense Connection Strategy
HAT_FIR	30.77	-	SwinFIR: Revisiting the SwinIR with Fast Fourier Convolution and Improved Training for Image Super-Resolution
CPAT	30.52	0.8923	Channel-Partitioned Windowed Attention And Frequency Learning for Single Image Super-Resolution	-

0 of 22 row(s) selected.