3D Reconstruction and Self-calibration based on Binocular Stereo Vision

Abstract

A 3D reconstruction technique from stereo images that requires minimal intervention from the user has been developed. The reconstruction problem consists of three steps of estimating specific geometry groups. The first step is estimating the epipolar geometry that exists between the stereo image pairs which includes feature matching in both images. The second is estimating the affine geometry, a process to find a special plane in the projective space by means of vanishing points. The third step, which includes camera self-calibration, is obtaining a metric geometry from which a 3D model of the scene could be obtained. The major advantage of this method is that the stereo images do not need to be calibrated for reconstruction. The results of camera calibration and reconstruction have shown the possibility of obtaining a 3D model directly from features in the images.

Realistic 3D model generation of a real product based on 2D-3D registration

“…One simple way is selecting a set of points in each image of which corresponding points on the surface are known to solve the 2D-3D relationship [2][3]. From these pairs of points, the relationship between the image view and the 3D object often called camera transformation can be directly determined using standard camera calibration method [4][5]. However, the object should have geometric feature points which can be easily located in the image to find these pairs of points.…”