Comparing our method with a state-of-the-art single image denoising [Dabov et al.] and an existing multi-view stereo denoising method [Vaish et al.] on synthetic Gaussian noise.
In Matlab, NoisyImage = CleanImage + 50*randn(H,W,3).
Comparison between different denoising approaches
![]() |
One of 25 noisy input images, PSNR=16.0809 |
![]() |
![]() |
Single image denoising [Dabov et al.], PSNR=31.154 |
25-view stereo denoising [Vaish et al.], PSNR=21.062 |
![]() |
![]() |
Our 25-view denoising, PSNR=33.781 |
Ground truth |
Comparison between different denoising approaches (Insets)
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
Noisy Patch |
Single image denoising [Dabov et al.] |
25-view stereo denoising [Vaish et al.] |
Our results |
Ground Truth |