Deblurring On Hide Trained On Gopro

评估指标

PSNR (sRGB)

Params (M)

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
MAXIM	32.83	22.2	0.956	MAXIM: Multi-Axis MLP for Image Processing
AdaRevD	32.35	-	0.953	AdaRevD: Adaptive Patch Exiting Reversible Decoder Pushes the Limit of Image Deblurring
ID-Blau (FFTformer)	31.94	-	0.949	ID-Blau: Image Deblurring by Implicit Diffusion-based reBLurring AUgmentation
CAPTNet	31.86	-	0.949	Prompt-based Ingredient-Oriented All-in-One Image Restoration
LoFormer	31.86	-	0.949	LoFormer: Local Frequency Transformer for Image Deblurring
DeepRFT+	31.66	-	0.946	Intriguing Findings of Frequency Selection for Image Deblurring
ID-Blau (Restormer)	31.66	-	0.947	ID-Blau: Image Deblurring by Implicit Diffusion-based reBLurring AUgmentation
FFTformer	31.62	16.6	0.9455	Efficient Frequency Domain-based Transformers for High-Quality Image Deblurring
LaKDNet	31.58	37.5	0.946	Revisiting Image Deblurring with an Efficient ConvNet
ID-Blau (Stripformer)	31.50	-	0.944	ID-Blau: Image Deblurring by Implicit Diffusion-based reBLurring AUgmentation
Restormer-TLC	31.49	26.13	0.945	Improving Image Restoration by Revisiting Global Information Aggregation
DeblurDiNAT-L	31.47	16.1	0.944	DeblurDiNAT: A Compact Model with Exceptional Generalization and Visual Fidelity on Unseen Domains
Restormer	31.22	26.13	0.942	Restormer: Efficient Transformer for High-Resolution Image Restoration
MPRNet-TLC	31.19	20.1	0.942	Improving Image Restoration by Revisiting Global Information Aggregation
Stripformer	31.03	-	0.94	Stripformer: Strip Transformer for Fast Image Deblurring
MPRNet	30.96	20.1	0.939	Multi-Stage Progressive Image Restoration
Uformer-B	30.83	50.88	0.952	Uformer: A General U-Shaped Transformer for Image Restoration
BANet	30.16	-	0.93	BANet: Blur-aware Attention Networks for Dynamic Scene Deblurring
Suin et al	29.98	-	0.930	-	-
MT-RNN	29.15	2.6	0.918	Multi-Temporal Recurrent Neural Networks For Progressive Non-Uniform Single Image Deblurring With Incremental Temporal Training

0 of 25 row(s) selected.