Smartphones as 3D Imagers: A Novel Hardware Configuration with Structure from Motion Techniques

Abstract

Embedded cameras in smartphones possess a huge potential to solve many critical problems in a diverse field of computer vision. Some examples are robot navigation, identity verification, construction innovation, asset management and digital twins, manufacturing and factory automation, logistics and supply chains, and environmental conservation. The possibility to do 3D imaging with smartphone cameras is an exciting domain whose importance is recognized by many, including tech giants such as Apple (e.g., the Measure app), Lenovo, and Samsung. Key smartphone-based 3D imaging innovations are taking place for indoor mapping (for construction progress monitoring, facility management, and interior design), disabled people assistance, and robot navigation based on artificial intelligence techniques. The fundamental purpose of my work is to develop a smart solution to map the 3D indoor environment to solve the scale problem. To measure the depth of the object, I attached a laser to the smartphone and used SFM for 3D image processing. The algorithm provides the dense point cloud and by analyzing the close points, the reconstruction of the environment in 3D view is shown in the results.