Semi-Sparsity for Smoothing Filters

TL;DR

This paper introduces a semi-sparsity smoothing algorithm that effectively handles both sparse features and non-sparse polynomial surfaces using a novel optimization framework, enhancing signal and image processing tasks.

Contribution

The paper presents a new semi-sparsity prior and an efficient optimization method for feature-aware smoothing, extending sparsity-based techniques to more general surface types.

Findings

Effective in preserving edges and singularities while smoothing polynomial surfaces.

Utilizes half-quadratic splitting with FFTs for fast computation.

Demonstrates versatility across various signal and image processing applications.

Abstract

In this paper, we propose an interesting semi-sparsity smoothing algorithm based on a novel sparsity-inducing optimization framework. This method is derived from the multiple observations that semi-sparsity prior knowledge is more universally applicable, especially in areas where sparsity is not fully admitted, such as polynomial-smoothing surfaces. We illustrate that this semi-sparsity can be identified into a generalized $L_0$-norm minimization in higher-order gradient domains, thereby giving rise to a new "feature-aware" filtering method with a powerful simultaneous-fitting ability in both sparse features (singularities and sharpening edges) and non-sparse regions (polynomial-smoothing surfaces). Notice that a direct solver is always unavailable due to the non-convexity and combinatorial nature of $L_0$-norm minimization. Instead, we solve the model based on an efficient half-quadratic splitting minimization with fast Fourier transforms (FFTs) for acceleration. We finally demonstrate its versatility and many benefits to a series of signal/image processing and computer vision applications.