The monocular model-based 3D pose tracking

Image acquisition devices can usually only get two-dimensional plane images. It has always been research focus in the field of computer vision that how to quickly and accurately recover motion parameters of objects in space from two-dimensional plane images. In this paper, the 3D pose tracking algorithm of monocular model-based is studied. This paper firstly uses the technology of binocular stereo vision to achieve Three-dimensional reconstruction of the objects and uses a two-step method based on the reweighted brightness depth change constraint equation to solve pose on the frame. The paper adopts the brightness change constraint equation to get the coarse value of 3D pose, and then proposes the reweighted brightness depth method based on Huber function to get the exact value of 3D pose. The paper also presents the multi-frame registration method of view to solve the registration failure problem and eliminate the accumulated error. Experiments results prove a great advantage of real-time and accuracy compared with other systems.

[1]  Alexander Zelinsky,et al.  Real-time stereo tracking for head pose and gaze estimation , 2000, Proceedings Fourth IEEE International Conference on Automatic Face and Gesture Recognition (Cat. No. PR00580).

[2]  Vincent Lepetit,et al.  Monocular Model-based 3d Tracking of Rigid Objects (Foundations and Trends in Computer Graphics and Vision(R)) , 2005 .

[3]  Pat Langley,et al.  An Analysis of Bayesian Classifiers , 1992, AAAI.

[4]  In-So Kweon,et al.  Stereo Matching with the Distinctive Similarity Measure , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[5]  Qiang Ji,et al.  Real-Time Eye, Gaze, and Face Pose Tracking for Monitoring Driver Vigilance , 2002, Real Time Imaging.

[6]  Dorin Comaniciu,et al.  Mean Shift: A Robust Approach Toward Feature Space Analysis , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  Hongbin Zha,et al.  Affine correspondence based head pose estimation for a sequence of images by using a 3D model , 2004, Sixth IEEE International Conference on Automatic Face and Gesture Recognition, 2004. Proceedings..

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

[9]  Robert T. Collins,et al.  Mean-shift blob tracking through scale space , 2003, 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings..

[10]  Andrew Calway,et al.  Using affine correspondence to estimate 3-D facial pose , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).

[11]  Ling Chen,et al.  Large Head Movement Tracking Using Scale Invariant View-Based Appearance Model , 2007, ACIVS.

[12]  Qiang Ji,et al.  3D Face pose estimation and tracking from a monocular camera , 2002, Image Vis. Comput..

[13]  Trevor Darrell,et al.  3D pose tracking with linear depth and brightness constraints , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[14]  Mohinder S. Grewal,et al.  Kalman Filtering: Theory and Practice Using MATLAB , 2001 .

[15]  Takeo Kanade,et al.  An Iterative Image Registration Technique with an Application to Stereo Vision , 1981, IJCAI.