Robust Piecewise-Constant Smoothing: M-Smoother Revisited

 Linchao Bao         Qingxiong Yang        
  City University of Hong Kong          

Abstract    A robust estimator, namely M-smoother, for piecewise-constant smoothing is revisited in this paper. Starting from its generalized formulation, we propose a numerical scheme/framework for solving it via a series of weighted-average filtering (e.g., box filtering, Gaussian filtering, bilateral filtering, and guided filtering). Because of the equivalence between M-smoother and local-histogram-based filters (such as median filter and mode filter), the proposed framework enables fast approximation of histogram filters via a number of box filtering or Gaussian filtering. In addition, high-quality piecewise-constant smoothing can be achieved via a number of bilateral filtering or guided filtering integrated in the proposed framework. Experiments on depth map denoising show the effectiveness of our framework.


1. Win32 Binary Executable with GUI [zip (6.3MB) | README | HOWTO (video, 1.2MB)]
2. Full Experimental Validation PSNR Data for Fig 1 [Microsoft Excel data (.xlsx, 174KB) | Matlab data (.mat, 63KB) | Matlab plot code]


The materials in this page are supplementary materials for our paper at arXiv:1410.7580[cs.CV].