RSBlur

Introduced 2022-02-17

The RSBlur dataset provides pairs of real and synthetic blurred images with ground truth sharp images. The dataset enables the evaluation of deblurring methods and blur synthesis methods on real-world blurred images. Training, validation, and test sets consist of 8,878, 1,120, and 3,360 blurred images, respectively.

Benchmarks

10-shot image generation/Average PSNR 10-shot image generation/SSIM 2D Classification/Average PSNR 2D Classification/SSIM Blind Image Deblurring/Average PSNR Blind Image Deblurring/SSIM Deblurring/Average PSNR Deblurring/SSIM

Related Benchmarks

RSBlur (trained on synthetic)/10-shot image generation/Average PSNR RSBlur (trained on synthetic)/2D Classification/Average PSNR RSBlur (trained on synthetic)/Blind Image Deblurring/Average PSNR RSBlur (trained on synthetic)/Deblurring/Average PSNR