Search Results for author:
Found 6 papers, 4 papers with code
Generalization in diffusion models arises from geometry-adaptive harmonic representations
Finally, we show that when trained on regular image classes for which the optimal basis is known to be geometry-adaptive and harmonic, the denoising performance of the networks is near-optimal.
Learning multi-scale local conditional probability models of images
We instantiate this model using convolutional neural networks (CNNs) with local receptive fields, which enforce both the stationarity and Markov properties.
Stochastic Solutions for Linear Inverse Problems using the Prior Implicit in a Denoiser
Two recent lines of work – Denoising Score Matching and Plug-and-Play – propose methodologies for drawing samples from this implicit prior and using it to solve inverse problems, respectively.
Solving Linear Inverse Problems Using the Prior Implicit in a Denoiser
Here, we develop a robust and general methodology for making use of this implicit prior.
Interpretable and robust blind image denoising with bias-free convolutional neural networks
Here, however, we show that bias terms used in most CNNs (additive constants, including those used for batch normalization) interfere with the interpretability of these networks, do not help performance, and in fact prevent generalization of performance to noise levels not including in the training data.
Robust and interpretable blind image denoising via bias-free convolutional neural networks
In contrast, a bias-free architecture -- obtained by removing the constant terms in every layer of the network, including those used for batch normalization-- generalizes robustly across noise levels, while preserving state-of-the-art performance within the training range.
