Fast Unsupervised Tensor Restoration via Low-rank Deconvolution
This provides an efficient alternative to deep learning methods for signal restoration tasks, though it appears incremental as an extension of an existing analytical model.
The authors tackled signal restoration problems like image denoising and video enhancement by extending Low-rank Deconvolution with differential regularization, achieving considerable performance improvements while maintaining low computational cost.
Low-rank Deconvolution (LRD) has appeared as a new multi-dimensional representation model that enjoys important efficiency and flexibility properties. In this work we ask ourselves if this analytical model can compete against Deep Learning (DL) frameworks like Deep Image Prior (DIP) or Blind-Spot Networks (BSN) and other classical methods in the task of signal restoration. More specifically, we propose to extend LRD with differential regularization. This approach allows us to easily incorporate Total Variation (TV) and integral priors to the formulation leading to considerable performance tested on signal restoration tasks such image denoising and video enhancement, and at the same time benefiting from its small computational cost.