Iteratively Reweighted Graph Cut for Multi-label MRFs with Non-convex Priors

Abstract

While widely acknowledged as highly effective in computer vision, multi-label MRFs with non-convex priors are difficult to optimize. To tackle this, we introduce an algorithm that iteratively approximates the original energy with an appropriately weighted surrogate energy that is easier to minimize. Our algorithm guarantees that the original energy decreases at each iteration. In particular, we consider the scenario where the global minimizer of the weighted surrogate energy can be obtained by a multi-label graph cut algorithm, and show that our algorithm then lets us handle of large variety of non-convex priors. We demonstrate the benefits of our method over state-of-the-art MRF energy minimization techniques on stereo and inpainting problems.

9 pages, 5 figures and 6 tables

Cited by in corpus (8)

• Deeply Supervised Depth Map Super-Resolution as Novel View Synthesis
• A new stereo formulation not using pixel and disparity models
• Generalized Range Moves
• Fast and Differentiable Message Passing on Pairwise Markov Random Fields
• Memory Efficient Max Flow for Multi-label Submodular MRFs
• cvpaper.challenge in 2015 - A review of CVPR2015 and DeepSurvey
• Deep Depth Super-Resolution : Learning Depth Super-Resolution using Deep Convolutional Neural Network
• cvpaper.challenge in 2016: Futuristic Computer Vision through 1,600 Papers Survey