Structure-preserving image smoothing with semantic cues

Abstract

The purpose of image smoothing is to smooth out low-contrast textures while preserving meaningful structures. Although this problem has been studied for decades, it still leaves a lot of space to improve. Recently, learning-based edge detectors have superior performance to traditional manually-designed detectors. Based on the edge detection technique, we present a novel optimization-based image smoothing model combining semantic prior and perform L 0 gradient minimization recursively in our framework to refine the result. Our framework combines the advantage of the state-of-the-art edge detector and the ability of L 0 gradient minimization for structure-preserving image smoothing. Moreover, we employ a large number of real-world images and perform various experiments to evaluate our algorithm. Experimental results show that our algorithm outperforms state-of-the-art algorithms, especially in extracting subjectively-meaningful structures.

An 𝓵p-nonconvex regularization method for image smoothing

“…This iteration can be implemented at a low computational cost, since the resulting matrix in (15) is tridiagonal.…”

“…In [14], Zhao et al proposed a framework called local activity-driven relative total variation, which can efficiently extract structural information. In [15], Chen et al presented a new structure-preserving smoothing model that incorporates the semantic prior and performs 𝓵𝓵0 gradient minimization recursively to refine results. In [16], Wang et al introduced a new method for structure-preserving image processing based on a mathematical derivation of the strict data-specific anisotropic Mexican hat wavelets (DAM).…”