Graph cut based multiple view segmentation for 3d reconstruction

In this paper we propose a novel framework for effi-ciently extracting foreground objects in so called short-baseline image sequences. We apply the obtained segmen-tation to improve subsequent 3D reconstruction results. Es-sentially, our framework combines a graph cut based opti-mization algorithm with an intuitive user interface. At first a meanshift segmentation algorithm partitions each image of the sequence into a certain number of regions. Additionally we provide an intelligent graphical user interface for easy specification of foreground as well as background regions across all images of the sequence. Within the graph cut op-timization algorithm we define new energy terms to increase the robustness and to keep the segmentation of the fore-ground object coherent across all images of the sequence. Finally, a refined graph cut segmentation and several ad-justment operations allow an accurate and effective fore-ground extraction. The obtained results are demonstrated on several real world data sets. 1