Adaptive compressive-sensing of 3D point clouds

In this paper we propose a mechanism for adaptive compressive-sensing of 3D point cloud. One problem with conventional compressive-sensing algorithm is a-priori fixed compression rate. In this work, we show that there is a correlation between sparsity of point cloud and number of edge points. The algorithm uses this information to estimate sparsity of input point cloud and adaptively change compression rate to achieve the same reconstruction error.