Abstract
Today, Convolutional Neural Networks (CNNs) are the leading method for image denoising. They are traditionally trained on pairs of images, which are often hard to obtain for practical applications. This motivates self-supervised training methods such as Noise2Void~(N2V) that operate on single noisy images. Self-supervised methods are, unfortunately, not competitive with models trained on image pairs. Here, we present 'Probabilistic Noise2Void' (PN2V), a method to train CNNs to predict per-pixel intensity distributions. Combining these with a suitable description of the noise, we obtain a complete probabilistic model for the noisy observations and true signal in every pixel. We evaluate PN2V on publicly available microscopy datasets, under a broad range of noise regimes, and achieve competitive results with respect to supervised state-of-the-art methods.
Highlights
Image restoration is the problem of reconstructing an image from a corrupted version of itself
In this work we show that MMSEPN2V consistently outperforms other self-supervised methods, and in many cases, leads to results that are competitive even with supervised state-of-the-art content-aware image restoration (CARE) networks
We have introduced Probabilistic Noise2Void (PN2V), a fully probabilistic approach extending self-supervised CARE training
Summary
Image restoration is the problem of reconstructing an image from a corrupted version of itself. Self-supervised training methods, such as Noise2Void (N2V) (Krull et al, 2019), are a promising alternative, as they operate exclusively on single noisy images (Batson and Royer, 2019; Krull et al, 2019; Laine et al, 2019). This is enabled by excluding/masking the center (blind-spot) of the network’s receptive fields. It does not depend on the content of the images that are to be denoised
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
