Real-time coarse-to-fine topologically preserving segmentation
| Autor: | Raquel Urtasun, Marko Boben, Sanja Fidler, Jian Yao |
|---|---|
| Rok vydání: | 2015 |
| Předmět: |
Markov random field
Segmentation-based object categorization business.industry Inference Scale-space segmentation Boundary (topology) Pattern recognition Topology (electrical circuits) Image segmentation Computer Science::Computer Vision and Pattern Recognition Segmentation Artificial intelligence business Mathematics |
| Zdroj: | CVPR |
| DOI: | 10.1109/cvpr.2015.7298913 |
| Popis: | 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. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|