Romps, Principles of stereo reconstruction of aerial objects using stationary cameras, RSL, 2024
Paper
Description
Our group began operating stereo cameras to study clouds in 2011, and this paper is the guide I wish we had back then. Stereo cameras are not complicated: they are just two or more regular cameras. But to combine their images to perform a stereo reconstruction, the equations and various conventions can be confusing. In this paper, stereo reconstruction is stripped down to its basic elements and presented using conventions tailored to applications in atmospheric science. In addition, the resulting equations are used to illustrate the high sensitivity of reconstructed cloud positions to errors in the cameras' Euler angles. If you are interested in performing your first stereo reconstruction, you will find here a self-contained guide to doing so using distortion-corrected images from a pair of calibrated, stationary cameras, as well as a demonstration of the need for high accuracy in the measurement of camera properties and orientations.
Images from the (left) left and (right) right stereo cameras taken at 15:15:20 UTC on August 27, 2022 during the TRACER campaign. Colored arrows have been added to draw attention to a sample of five feature points (among hundreds) matched between the two images.