Abstract

Omnidirectional catadioptric systems are popular in many fields, such as robotics, surveillance, telepresence and virtual reality, thanks to their large field of view. For 3D scene reconstruction in a single shot, usually two different catadioptric cameras are needed. More cameras may contribute to better reconstruction while a larger mounting space and higher power cost are required. In this paper, a single camera multi-mirror catadioptric system with a vertical and horizontal baseline structure is proposed. It features the realization of multiple central or non-central omnidirectional stereo pairs in a compact manner. To make the 3D reconstruction process general and adaptive to various types of system configurations, a flexible calibration and reconstruction algorithm pipeline is presented. The algorithm features the approximation of the system into multiple central sub-cameras and carries out stereo matching in a spherical representation. In addition, an effective 3D point cloud fusion algorithm is proposed to optimize the reconstruction results from multiple stereo pairs. Experiments carried out with synthetic and real data verified the feasibility and effectiveness of our system. In particular, with the real data, when compared with the results of a single stereo pair, the fused reconstruction results of the multiple stereo pairs obtained 30% and 15% improvement corresponding to the planarity and angle error respectively.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call