Parameter-free structure-texture image decomposition by unrolling
This work addresses image decomposition for computer vision applications, but it is incremental as it builds on existing unrolling techniques.
The authors tackled the structure-texture image decomposition problem by proposing LPR-NET, a parameter-free neural network based on unrolling the Low Patch Rank model, which learns parameters from data and is computationally faster while achieving qualitatively similar results to traditional methods, with numerical experiments showing good generalization to natural images despite training on synthetic data.
In this work, we propose a parameter-free and efficient method to tackle the structure-texture image decomposition problem. In particular, we present a neural network LPR-NET based on the unrolling of the Low Patch Rank model. On the one hand, this allows us to automatically learn parameters from data, and on the other hand to be computationally faster while obtaining qualitatively similar results compared to traditional iterative model-based methods. Moreover, despite being trained on synthetic images, numerical experiments show the ability of our network to generalize well when applied to natural images.