Incremental Learning Method for Unified Camera Calibration
The camera model could be approximated by a set of linear models defined on a set of local receptive fields regions. Camera calibration could then be a learning procedure to evolve the size and shape of every receptive field as well as parameters of the associated linear model. For a multi-camera system, its unified model is obtained from a fusion procedure integrated with all linear models weighted by their corresponding approximation measurements. The 3-D measurements of the multi-camera vision system are produced from a weighted regression fusion on all receptive fields of cameras. The resultant calibration model of a multi-camera system is expected to have higher accuracy than either of them. Simulation and experiment results illustrate effectiveness and properties of the proposed method. Comparisons with the Tsai’s method are also provided to exhibit advantages of the method.
KeywordsWeighted Regression Corner Point Reconstruction Error Camera Calibration Incremental Learn
Unable to display preview. Download preview PDF.
- 1.Hashimoto, H., Lee, J.H., Ando, N.: Self-Identification of Distributed Intelligent Networked Device in Intelligent Space. In: Proc. of IEEE Inter. Conf. on Robotics and Automation, pp. 4172–4177 (2003)Google Scholar
- 3.Porikli, F., Divakaran, A.: Multi-Camera Calibration, Object Tracking and Query Generation. In: Proc. of Inter. Conf. on Multimedia and Expo, pp. 653–656 (2003)Google Scholar
- 5.Ahmed, M.T., Farag, A.A.: A Neural Optimization Framework for Zoom Lens Camera Calibration. In: Proc. of IEEE Conf. on Computer Vision and Pattern Recognition, vol. 1, pp. 403–409 (2000)Google Scholar
- 6.Do, Y.T.: Application of Neural Network for Stereo-Camera Calibration. In: Proc. of the IEEE/RSJ Inter. Conf. on Intelligent Robots and Systems, pp. 2719–2722 (1999)Google Scholar
- 7.Jun, J., Kim, C.: Robust Camera Calibration Using Neural Network. In: Proc. of IEEE Region 10 Conference, vol. 1, pp. 694–697 (1999)Google Scholar