A Non-negative Matrix Factorization Framework for Combining Multiple Image Segmentations ; CU-CS-1050-08

Segmentation, or partitioning images into internally homogeneous regions, is an important first step in many Computer Vision tasks. In this paper, we attack the segmentation problem using an ensemble of low cost image segmentations. These segmentations are reconciled by applying recent techniques from the consensus clustering literature which exploit a Non-negative Matrix Factorization (NMF) framework. We describe extensions to these methods that scale them for large images and incorporate smoothness constraints. This framework allows us to uniformly combine segmentations from different algorithms or feature modalities, while avoiding significant parameter tuning for the specific image being segmented. We demonstrate that combining multiple “naive” image segmentations derived from k-means clustering compares favorably with more advanced Mean Shift and Efficient Graph Based Segmentation algorithms. The approach is evaluated on the Berkeley image segmentation dataset.