Image Super Resolution On Set5 2X Upscaling

评估指标

PSNR

SSIM

评测结果

各个模型在此基准测试上的表现结果

			Paper Title	Repository
DRCT-L	39.14	0.9658	DRCT: Saving Image Super-resolution away from Information Bottleneck
HMA†	38.95	0.9649	HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution
HAT-L	38.91	0.9646	Activating More Pixels in Image Super-Resolution Transformer
Hi-IR-L	38.87	0.9663	Hierarchical Information Flow for Generalized Efficient Image Restoration	-
HAT_FIR	38.74	-	SwinFIR: Revisiting the SwinIR with Fast Fourier Convolution and Improved Training for Image Super-Resolution
HAT	38.73	0.9637	Activating More Pixels in Image Super-Resolution Transformer
CPAT+	38.72	0.9635	Channel-Partitioned Windowed Attention And Frequency Learning for Single Image Super-Resolution	-
DRCT	38.72	0.9646	DRCT: Saving Image Super-resolution away from Information Bottleneck
CPAT	38.68	0.9633	Channel-Partitioned Windowed Attention And Frequency Learning for Single Image Super-Resolution	-
SwinFIR	38.65	0.9633	SwinFIR: Revisiting the SwinIR with Fast Fourier Convolution and Improved Training for Image Super-Resolution
EDT-B	38.63	0.9632	On Efficient Transformer-Based Image Pre-training for Low-Level Vision
MaIR+	38.62	0.963	MaIR: A Locality- and Continuity-Preserving Mamba for Image Restoration
MaIR	38.56	0.9628	MaIR: A Locality- and Continuity-Preserving Mamba for Image Restoration
DRLN+	38.34	0.9619	Densely Residual Laplacian Super-Resolution
LTE	38.33	-	Local Texture Estimator for Implicit Representation Function
HAN+	38.33	0.9299	Single Image Super-Resolution via a Holistic Attention Network
CSNLN	38.28	0.9616	Image Super-Resolution with Cross-Scale Non-Local Attention and Exhaustive Self-Exemplars Mining
FACD	38.242	-	Feature-domain Adaptive Contrastive Distillation for Efficient Single Image Super-Resolution	-
PMRN+	38.22	0.9612	Sequential Hierarchical Learning with Distribution Transformation for Image Super-Resolution	-
SwinOIR	38.21	0.9614	Resolution Enhancement Processing on Low Quality Images Using Swin Transformer Based on Interval Dense Connection Strategy

0 of 41 row(s) selected.