首页 | 本学科首页   官方微博 | 高级检索  
     检索      


Multi-image 3D reconstruction data evaluation
Institution:1. Institute for Language and Speech Processing, Multimedia Department, Research Centre “Athena”, PO BOX 159, 67100 Xanthi, Greece;2. University of Ljubljana, Faculty of Civil and Geodetic engineering, 1000 Ljubljana, Slovenia;3. Democritus University of Thrace, Faculty of Electrical and Computer Engineering, 67100 Xanthi, Greece;1. School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, China;2. School of Educational Information Technology, Central China Normal University, Wuhan 430079, China;3. Institute of Oceanographic Instrumentation, Shandong Academy of Sciences, Qingdao 266001, China;1. Center for Cellular Imaging and NanoAnalytics, Biozentrum, University of Basel, Mattenstrasse 26, 4058 Basel, Switzerland;2. Department of BioNanoscience, Delft University of Technology, Van der Waalsweg 8, 2628 CH Delft, The Netherlands;3. Department of Pharmacology, Case Western Reserve University, 10900 Euclid Avenue, Wood Bldg 321D, Cleveland, OH 44106-4965, USA
Abstract:A number of software solutions based on the Structure-From-Motion (SFM) and Dense Multi-View 3D Reconstruction (DMVR) algorithms have been made recently available. They allow the production of high quality 3D models by using unordered image collections that depict a scene or an object from different viewpoints. In this work, we question the quality of the data produced by a commercial SFM-DMVR software. An Ottoman monument located in the region of Xanthi, Greece has been selected as a case study. We attempted to quantify the quality of the SFM-DMVR data in relation to the data produced by a Time-of-Flight terrestrial 3D range scanner. We have implemented a number of comparisons between different parts of the monument in order to assess the mesh deviations and the reconstruction's accuracy. In order to further ensure the validity of our evaluation phase, we performed additional distance measurements between feature points on the monument's surface by using a total station and empirical measurements. The applicability of the SFM-DMVR method was questioned by creating a complete 3D digital replica of the monument.
Keywords:
本文献已被 ScienceDirect 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号