Factorization of canonic homographies for camera calibration and scene modeling

This paper proposes a novel framework to calibrate cameras and model a scene simultaneously using both self-calibration constraints and geometric information on parallelograms. The proposed method is a factorization-based approach solving the problem by decomposing a measurement matrix into camera and parallelogram parameters. Since the factorization-based approach recovers all camera poses simultaneously, the consistency of rigid transformations among cameras can be ensured. This paper also suggests the strategy dealing with the missing data, which is usually an obstacle to factorization-based approaches. The proposed method is more useful than the previous factorization-based approach using parallelepipeds because parallelograms are more basic scene elements. Moreover, this paper describes the unifying method using the constraints from parallelograms and parallelepipeds simultaneously in the proposed framework. The results of the experiments with synthetic data and real outdoor images are presented to demonstrate the feasibility of the proposed method.