Image Denoising On Sidd

评估指标

PSNR (sRGB)

SSIM (sRGB)

评测结果

各个模型在此基准测试上的表现结果

			Paper Title	Repository
NBNet	-	0.973	NBNet: Noise Basis Learning for Image Denoising with Subspace Projection
KBNet	40.35	0.972	KBNet: Kernel Basis Network for Image Restoration
CGNet	40.39	0.964	CascadedGaze: Efficiency in Global Context Extraction for Image Restoration
SSAMAN	40.08	0.962	Single Stage Adaptive Multi-Attention Network for Image Restoration	-
AKDT	39.70	0.961	AKDT: Adaptive Kernel Dilation Transformer for Effective Image Denoising	-
NAFNet	40.30	0.961	Simple Baselines for Image Restoration
MAXIM-3S	39.96	0.960	MAXIM: Multi-Axis MLP for Image Processing
Restormer	40.02	0.960	Restormer: Efficient Transformer for High-Resolution Image Restoration
Uformer-B	39.89	0.960	Uformer: A General U-Shaped Transformer for Image Restoration
PNGAN	40.07	0.960	Learning to Generate Realistic Noisy Images via Pixel-level Noise-aware Adversarial Training
MIRNet	39.72	0.959	Learning Enriched Features for Real Image Restoration and Enhancement
SRMNet	39.72	0.959	Selective Residual M-Net for Real Image Denoising
HINet	39.99	0.958	HINet: Half Instance Normalization Network for Image Restoration
MPRNet	39.71	0.958	Multi-Stage Progressive Image Restoration
DRANet	-	0.957	Dual Residual Attention Network for Image Denoising
DANet+	39.47	0.957	Dual Adversarial Network: Toward Real-world Noise Removal and Noise Generation
SADNet	39.46	0.957	Spatial-Adaptive Network for Single Image Denoising
CycleISP	39.52	0.957	CycleISP: Real Image Restoration via Improved Data Synthesis
VDN	39.28	0.956	Variational Denoising Network: Toward Blind Noise Modeling and Removal
AINDNet	38.95	0.952	Transfer Learning from Synthetic to Real-Noise Denoising with Adaptive Instance Normalization

0 of 22 row(s) selected.