Image Super Resolution On Set14 3X 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

模型名称	PSNR	SSIM	Paper Title	Repository
HMA†	31.47	0.8585	HMANet: Hybrid Multi-Axis Aggregation Network for Image Super-Resolution
Hi-IR-L	31.55	0.8616	Hierarchical Information Flow for Generalized Efficient Image Restoration	-
HAT	31.33	0.8576	Activating More Pixels in Image Super-Resolution Transformer
HAN+	30.79	0.8487	Single Image Super-Resolution via a Holistic Attention Network
SwinOIR	30.65	0.8493	Resolution Enhancement Processing on Low Quality Images Using Swin Transformer Based on Interval Dense Connection Strategy
SwinFIR	31.24	0.8566	SwinFIR: Revisiting the SwinIR with Fast Fourier Convolution and Improved Training for Image Super-Resolution
CSNLN	30.66	0.8482	Image Super-Resolution with Cross-Scale Non-Local Attention and Exhaustive Self-Exemplars Mining
ML-CrAIST	30.39	0.8488	ML-CrAIST: Multi-scale Low-high Frequency Information-based Cross black Attention with Image Super-resolving Transformer
HAT-L	31.47	0.8584	Activating More Pixels in Image Super-Resolution Transformer
LTE	30.8	-	Local Texture Estimator for Implicit Representation Function
CPAT	31.15	0.8557	Channel-Partitioned Windowed Attention And Frequency Learning for Single Image Super-Resolution	-
RED30	29.61	0.8341	Image Restoration Using Convolutional Auto-encoders with Symmetric Skip Connections
CPAT+	31.19	0.8559	Channel-Partitioned Windowed Attention And Frequency Learning for Single Image Super-Resolution	-
LCSCNet	29.87	-	LCSCNet: Linear Compressing Based Skip-Connecting Network for Image Super-Resolution
SRFBN	30.1	-	Feedback Network for Image Super-Resolution
Deep CNN Denoiser	27.72	-	Learning Deep CNN Denoiser Prior for Image Restoration
PMRN+	29.24	0.8087	Sequential Hierarchical Learning with Distribution Transformation for Image Super-Resolution	-
HAT_FIR	31.37	-	SwinFIR: Revisiting the SwinIR with Fast Fourier Convolution and Improved Training for Image Super-Resolution
IPT	30.85	-	Pre-Trained Image Processing Transformer
MWCNN	30.16	-	Multi-level Wavelet-CNN for Image Restoration

0 of 24 row(s) selected.