Blind Image Restoration without Prior Knowledge
This addresses the issue of performance decline in image restoration when degradation conditions vary, offering a more adaptable solution for computer vision applications, though it appears incremental as it builds on existing CNN topologies.
The paper tackles the problem of blind image restoration where existing methods require prior knowledge of degradation processes, and introduces the Self-Normalization Side-Chain (SNSC) that deduces degradation parameters from training data without such knowledge, demonstrating improved restoration performance across tasks.
Many image restoration techniques are highly dependent on the degradation used during training, and their performance declines significantly when applied to slightly different input. Blind and universal techniques attempt to mitigate this by producing a trained model that can adapt to varying conditions. However, blind techniques to date require prior knowledge of the degradation process, and assumptions regarding its parameter-space. In this paper we present the Self-Normalization Side-Chain (SCNC), a novel approach to blind universal restoration in which no prior knowledge of the degradation is needed. This module can be added to any existing CNN topology, and is trained along with the rest of the network in an end-to-end manner. The imaging parameters relevant to the task, as well as their dynamics, are deduced from the variety in the training data. We apply our solution to several image restoration tasks, and demonstrate that the SNSC encodes the degradation-parameters, improving restoration performance.