Interactive foreground segmentation and shape reconstruction from RGBD images

Three dimensional shape reconstruction of objects in complex scenes has important applications in intelligent transportation systems and medical image processing. This paper aims to segment and locate the foreground object through the interactive segmentation with multiple cues including saliency, depth and color. The desired foreground object is obtained by using the foreground and background information provided by saliency map and heatmap. Then, the multi-view point cloud information of the foreground object is recovered by depth information, and a multi-view point cloud registration algorithm based on color information is proposed. The three-dimensional shape model of the object is reconstructed through a multi-view point cloud registration algorithm combined with motion averaging and low-rank sparse matrix decomposition. The validity of the framework is empirically assessed with comparative experiments.