Comparison of compuer-vision-based photogrammetric software for reconsturcting virtual indigenous tribe

Yu Qi Lin, Sendo Wang

Research output: Contribution to conferencePaperpeer-review

Abstract

To help the young generation of urban indigenous peoples who is far from their home tribes to learn their traditional environmental knowledge, we proposed a novel concept called “Virtual Indigenous Tribe”. It is reconstructed by aerial images taken by an unmanned aerial vehicle (UAV) and panoramic images taken by a 360° dual-fish-eye camera. Benefit from the rapid development of computer vision algorithms and technologies, the ortho-rectified mosaic image and the 3D mesh model of an indigenous tribe can be comparably easier derived from aerial images. However, the precision of these computer-vision-based photogrammetric software are always challenged and their results will also differ widely by locations. Thus, we expect to compare the difference between these computer-vision-based photogrammetric software when reconstructing a virtual indigenous tribe. Pix4Dmapper and Agisoft Metashape Professional are the two computer-vision-based photogrammetric software chosen to compare with. We will first survey and compare the specifications of each software, such as hardware requirements, the difference between different versions, input data requirements and so on. Then, we generate 3D models using these software that all parameters are set to the same. We compare each software with the operating time, simplicity of operation, and the precision and accuracy of the results. Result shows that Pix4Dmapper is more effective and has higher degree of automation than Agisoft Metashape Professional. Therefore, we recommend Pix4Dmapper for teachers and people in tribes to reconstruct their Virtual Indigenous Tribes.

Original languageEnglish
Publication statusPublished - 2020
Event40th Asian Conference on Remote Sensing: Progress of Remote Sensing Technology for Smart Future, ACRS 2019 - Daejeon, Korea, Republic of
Duration: 2019 Oct 142019 Oct 18

Conference

Conference40th Asian Conference on Remote Sensing: Progress of Remote Sensing Technology for Smart Future, ACRS 2019
Country/TerritoryKorea, Republic of
CityDaejeon
Period2019/10/142019/10/18

Keywords

  • Computer Vision
  • Photogrammetry
  • Virtual Indigenous Tribe

ASJC Scopus subject areas

  • Information Systems

Fingerprint

Dive into the research topics of 'Comparison of compuer-vision-based photogrammetric software for reconsturcting virtual indigenous tribe'. Together they form a unique fingerprint.

Cite this