Real-time coarse-to-fine topologically preserving segmentation
Authors:
- 1 June 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10636919,p. 2947-2955
Abstract
In this paper, we tackle the problem of unsupervised segmentation in the form of superpixels. Our main emphasis is on speed and accuracy. We build on [31] to define the problem as a boundary and topology preserving Markov random field. We propose a coarse to fine optimization technique that speeds up inference in terms of the number of updates by an order of magnitude. Our approach is shown to outperform [31] while employing a single iteration. We evaluate and compare our approach to state-of-the-art superpixel algorithms on the BSD and KITTI benchmarks. Our approach significantly outperforms the baselines in the segmentation metrics and achieves the lowest error on the stereo task.Keywords
This publication has 23 references indexed in Scilit:
- Efficient Hierarchical Graph-Based Segmentation of RGBD Videos2014
- Fast and Accurate Large-Scale Stereo Reconstruction Using Variational Methods2013
- Robust Monocular Epipolar Flow Estimation2013
- Are we ready for autonomous driving? The KITTI vision benchmark suite2012
- Superpixels via pseudo-Boolean optimization2011
- Entropy rate superpixel segmentation2011
- Efficient hierarchical graph-based video segmentation2010
- TurboPixels: Fast Superpixels Using Geometric FlowsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2009
- Accurate and Efficient Stereo Processing by Semi-Global Matching and Mutual Information2005
- Efficient Graph-Based Image SegmentationInternational Journal of Computer Vision, 2004