Section: New Results

Local/Non-Local Noisy Image Deconvolution

Participants : Hicham Badri, Hussein Yahia.

Reference: [24] .

Image deconvolution is a standard step in many imaging applications. Sparse local regularization has shown to be fast but tends to over-smoothing images. On the other hand, non-local priors that manipulate similar patches produce better results but tend to be much slower. In this paper, we combine both local and non-local methods in one framework to offer both good quality image reconstruction and computational efficiency in the presence of noise. By studying the non-local singular values of the image patches, we show that the non-local patches tend to be much similar in the blurred version of the image. We thus use low-rank estimation to first estimate a blurred but noise-free image. Secondly, we show that this denoising step introduces outliers in the deconvotion model and propose anefficient optimization method to tackle this problem. Experiments show that the proposed method poduces comparable results to non-local methods while being more computationally efficient.

Figure 3. Motion estimation using the proposed method. From left to right: image sequences (2 images, at $t$ and $t+1$ respectively) the ground-truth and the estimated flow (errors, from left to right : MSE=0.063, AAE=3.562, EPE=0.100).