Highly Consistent Sequential Segmentation

Abstract

Abstract. This paper deals with segmentation of image sequences in an unsupervised manner with the goal of getting highly consistent segmentation results from frame-to-frame. We first introduce a segmentation method that uses results of the previous frame as initialization and significantly improves consistency in comparison to a single frame based approach. We also find correspondences between the segmented regions from one frame to the next to further increase consistency. This matching step is based on a modified version of an efficient partial shape matching method which allows identification of similar parts of regions despite topology changes like merges and splits. We use the identified matched parts to define a partial matching cost which is then used as input to pairwise graph matching. Experiments demonstrate that we can achieve highly consistent segmentations for diverse image sequences, even allowing to track manually initialized moving and static objects.

Edge-Weighted Centroid Voronoi Tessellation with Propagation of Consistency Constraint for 3D Grain Segmentation in Microscopic Superalloy Images

“…We select StreamGBH because it is a propagation-based method and it achieves state-of-the-art performance on video segmentation tasks [22], which is similar to the 3D grain image segmentation application. Another related work is the method in [8], however the authors have not released the implementation of this method.…”

“…In [8], the authors propagate the result of previous video frame as the initialization for segmenting the current frame. After a modified activecontour based segmentation, the algorithm further merges and splits segments according to the partial shape matching across video frames.…”