An Evolutionary-Based 3D Reconstruction Using an Uncalibrated Stereovision System

摘要

In this paper, we offer an original method for 3D shape reconstruction based on an uncalibrated stereovision system (USS). The USS is composed of five cameras located on an arc around the object to be analysed. The method operates in two steps. First, points of interest are detected in pairs of images acquired by two consecutive cameras of the USS and matched. Then, using evolutionary algorithms, we compute the transform matrix between the two images and the respective depths of the points of interest. The accuracy of the proposed method is validated through a comparison with the depth values obtained by a traditional method based on calibrated cameras.