Learning Deep Analysis Dictionaries -- Part II: Convolutional Dictionaries

TL;DR

This paper introduces DeepCAM, a deep convolutional analysis dictionary model that uses convolutional dictionaries for high-dimensional signals, demonstrating competitive performance in image super-resolution tasks.

Contribution

It proposes a novel multilayer convolutional analysis dictionary learning approach, extending previous models with convolutional dictionaries for improved efficiency and effectiveness.

Findings

Achieves performance comparable to state-of-the-art methods in super-resolution.

Demonstrates good generalization capabilities across tasks.

Introduces an efficient learning algorithm for convolutional analysis dictionaries.

Abstract

In this paper, we introduce a Deep Convolutional Analysis Dictionary Model (DeepCAM) by learning convolutional dictionaries instead of unstructured dictionaries as in the case of deep analysis dictionary model introduced in the companion paper. Convolutional dictionaries are more suitable for processing high-dimensional signals like for example images and have only a small number of free parameters. By exploiting the properties of a convolutional dictionary, we present an efficient convolutional analysis dictionary learning approach. A L-layer DeepCAM consists of L layers of convolutional analysis dictionary and element-wise soft-thresholding pairs and a single layer of convolutional synthesis dictionary. Similar to DeepAM, each convolutional analysis dictionary is composed of a convolutional Information Preserving Analysis Dictionary (IPAD) and a convolutional Clustering Analysis Dictionary (CAD). The IPAD and the CAD are learned using variations of the proposed learning algorithm. We demonstrate that DeepCAM is an effective multilayer convolutional model and, on single image super-resolution, achieves performance comparable with other methods while also showing good generalization capabilities.