Learning pseudo-contractive denoisers for inverse problems
This addresses a bottleneck in signal and image processing for researchers and practitioners by providing a more efficient convergence guarantee without sacrificing recovery quality, though it is incremental as it builds on existing denoiser frameworks.
The paper tackled the problem of enforcing Lipschitz constraints on deep denoisers for inverse problems, which can compromise performance, by introducing a novel training strategy that enforces a weaker pseudo-contractive constraint, resulting in superior performance compared to related denoisers with competitive visual effects and quantitative values.
Deep denoisers have shown excellent performance in solving inverse problems in signal and image processing. In order to guarantee the convergence, the denoiser needs to satisfy some Lipschitz conditions like non-expansiveness. However, enforcing such constraints inevitably compromises recovery performance. This paper introduces a novel training strategy that enforces a weaker constraint on the deep denoiser called pseudo-contractiveness. By studying the spectrum of the Jacobian matrix, relationships between different denoiser assumptions are revealed. Effective algorithms based on gradient descent and Ishikawa process are derived, and further assumptions of strict pseudo-contractiveness yield efficient algorithms using half-quadratic splitting and forward-backward splitting. The proposed algorithms theoretically converge strongly to a fixed point. A training strategy based on holomorphic transformation and functional calculi is proposed to enforce the pseudo-contractive denoiser assumption. Extensive experiments demonstrate superior performance of the pseudo-contractive denoiser compared to related denoisers. The proposed methods are competitive in terms of visual effects and quantitative values.