Towards automatic calibration of stereoscopic video systems

The last few years in the movie industry were marked by continuous growth in the popularity of 3D technology. 3D movies are nowadays a major draw for both cinema and TV audiences. The application of 3D technology in combination with modern ultra high-definition video cameras causes that today's movies are more impressive and realistic. However, it also makes their production more and more complex, demanding and expensive. One of the main issues encountered during recording of 3D movies is precise calibration of cameras working together in a 3D stereoscopic system. The calibration process should be performed very carefully to minimize all possible image distortions, because their correction in post-production is usually very long and costly, and sometimes even impossible. One of the most intuitive solution of this problem is calibration performed by a camera operator with using dedicated reference calibration boards. However, it is a long, iterative and subjective process. Taking into account that this process should be performed quite often, at least before every scene, the calibration greatly prolongs the preparation for the video recording and slows down the work on the movie set. All the aforementioned problems caused that the authors made an efforts aimed at automation of the calibration process. The paper presents the requirements and assumptions of the calibration method and first implementation of supporting software application. The main advantages of the presented method is significant reduction of the calibration time and a possibility of objective measurement of the calibration accuracy.

[1]  Manuel Mazo,et al.  Stereo Vision Tracking of Multiple Objects in Complex Indoor Environments , 2010, Sensors.

[2]  Robert S. Allison,et al.  Calibration for high-definition camera rigs with marker chessboard , 2012, 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops.

[3]  Peter Eisert,et al.  The Stereoscopic Analyzer — An image-based assistance tool for stereo shooting and 3D production , 2010, 2010 IEEE International Conference on Image Processing.

[4]  Paweł Pełczyński,et al.  Software Application for Calibration of Stereoscopic Camera Setups , 2012 .

[5]  Zhengyou Zhang,et al.  A Flexible New Technique for Camera Calibration , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[6]  Moritz Menze,et al.  USING STEREO VISION TO SUPPORT THE AUTOMATED ANALYSIS OF SURVEILLANCE VIDEOS , 2012 .

[7]  ZhangZhengyou A Flexible New Technique for Camera Calibration , 2000 .

[8]  Mehran Yazdi,et al.  Robust approach for people detection and tracking by stereo vision , 2014, 7'th International Symposium on Telecommunications (IST'2014).

[9]  Jorge Santolaria,et al.  A stereo-vision system to automate the manufacture of a semitrailer chassis , 2013 .

[10]  Arjun B. Krishnan,et al.  Intelligent Indoor Mobile Robot Navigation Using Stereo Vision , 2014, ArXiv.

[12]  Janne Heikkilä,et al.  Geometric Camera Calibration Using Circular Control Points , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[13]  Xin Yang,et al.  Multi-object detection and tracking by stereo vision , 2010, Pattern Recognit..