On-Line Dense Point Cloud Generation from Monocular Images with Scale Estimation

摘要: This paper introduces an approach for on-line marker-based three dimensional modeling with scale estimation and heightmap construction from monocular images. The presented system is also capable of off-line marker-less 3D reconstruction images increased detail. method designed the flexible use Unmaned Aerial Vehicle (UAV); this means that, despite being tested a Parrot AR.Drone 1.0, it easily portable to other more UAV models. followed was adaptation patch-based Multiview Stereo (PMVS) algorithm on line point cloud generation. achieved 1.05 processed per second average, slightly surpassing planed objective 1 image second. height error ranges between 1-1.5% manual marker detection 4-5% automatic detection, which seems accurate enough autonomous navigation path planning. As future work, tests better UAV, processing time reduction, map construction, indoor collaborative are planned.