Disaggregating Convolutional Dictionary Learning

In the past decade, many dictionary learning based disaggregation approaches have been proposed. The main shortcoming of dictionary learning is that it is not shift-invariant. This has been overcome in recent years through convolutional dictionary learning. Therefore, in this letter, we propose to develop a disaggregation technique based on convolutional dictionary learning. Comparison with all dictionary learning approaches and several state-of-the-art deep learning approaches on benchmark datasets show that our proposed technique improves over the rest.