Guided interactive image segmentation using machine learning and color based data set clustering

We present a novel approach that combines machine learning based interactive image segmentation with a two-stage clustering method to identify similarly colored images for efficient batch image segmentation by guided reuse of classifiers. The segmentation task is formulated as a supervised machine learning problem working on homogeneous groups of voxels termed supervoxels. Classifiers are interactively trained from sparse annotations in an iterative process of annotation refinement. Resulting models can be used for batch processing of previously unseen images. By clustering images into subsets of similar colorization, we identify a minimal set of prototype images and demonstrate that using only classifiers trained on these prototype images for their color-cluster significantly improves the average segmentation performance of batch processing. The presented methods are applicable for almost any image type and therefore represent a useful tool for image analysis tasks in general.