Multi-Object Geodesic Active Contours (MOGAC): A Parallel Sparse-Field Algorithm for Image Segmentation
View/ Open
Date
2012-02-28Author
Taylor, Russell
Kazhdan, Michael
Lucas, Blake
Metadata
Show full item recordAbstract
An important task for computer vision systems is to segment adjacent structures in images without producing gaps or overlaps. Multi-object Level Set Methods (MLSM) perform this task with the benefit of sub-pixel accuracy. However, current implementations of MLSM are not as computationally or memory efficient as their region growing and graph cut counterparts which lack sub-pixel accuracy. To address this performance gap, we present a novel parallel implementation of MLSM that leverages the sparse properties of the segmentation algorithm to minimize its memory footprint for multiple objects. The new method, Multi-Object Geodesic Active Contours (MOGAC), can represent N objects with just two functions: a label image and unsigned distance field. The time complexity of the algorithm is shown to be O((M^d)/P) for M^d pixels and P processing units in dimension d={2,3}, independent of the number of objects. Results are presented for 2D and 3D image segmentation problems.