Online Mesh Optimization for Large Scale KinectFusion Meshes

摘要

In this paper we present an extension of Large Scale Kinect Fusion to compute optimized triangle meshes on-the-fly by removing redundant triangles on planar surfaces. The optimization is integrated into the reconstruction pipeline to compute the optimized meshes asynchronously to the reconstruction process in real time. The computed reconstructions can be extracted directly without the need of any post processing.