Visual-model-based, real-time 3D pose tracking for autonomous navigation: methodology and experiments
- 255 Downloads
This paper presents a novel 3D-model-based computer-vision method for tracking the full six degree-of-freedom (dof) pose (position and orientation) of a rigid body, in real-time. The methodology has been targeted for autonomous navigation tasks, such as interception of or rendezvous with mobile targets. Tracking an object’s complete six-dof pose makes the proposed algorithm useful even when targets are not restricted to planar motion (e.g., flying or rough-terrain navigation). Tracking is achieved via a combination of textured model projection and optical flow. The main contribution of our work is the novel combination of optical flow with z-buffer depth information that is produced during model projection. This allows us to achieve six-dof tracking with a single camera.
A localized illumination normalization filter also has been developed in order to improve robustness to shading. Real-time operation is achieved using GPU-based filters and a new data-reduction algorithm based on colour-gradient redundancy, which was developed within the framework of our project. Colour-gradient redundancy is an important property of colour images, namely, that the gradients of all colour channels are generally aligned. Exploiting this property provides a threefold increase in speed. A processing rate of approximately 80 to 100 fps has been obtained in our work when utilizing synthetic and real target-motion sequences. Sub-pixel accuracies were obtained in tests performed under different lighting conditions.
KeywordsComputer vision Real-time object tracking Pose tracking Mobile-robot navigation
Unable to display preview. Download preview PDF.
- Advanced Multimedia Processing Lab. (2006, January). The self-reconfigurable camera array. Carnegie Mellon University. [Online]. Available: http://amp.ece.cmu.edu/projects/MobileCamArray/.
- Andrews, R., & Lovell, B. (2003). Color optical flow. In Workshop on digital image computing (Vol. 1, pp. 135–139) Brisbane, Australia, February 2003. Google Scholar
- ATI Technologies Inc. (2004, December). Radeon X800 graphics technology. [Online]. Available: http://www.ati.com/products/radeonx800/index.html.
- Barron, J., & Klette, R. (2002). Quantitative color optical flow. In 16th international conference on pattern recognition (Vol. 4, pp. 251–255), Quebec City, Canada, August 2002. Google Scholar
- Collins, G., & Dennis, L. A. (2000). A system for video surveillance and monitoring. In International conference on automated deduction (pp. 497–501), Pittsburgh, PA, June 2000. Google Scholar
- Comport, A., Marchand, E., & Chaumette, F. (2003). A real-time tracker for markerless augmented reality. In IEEE and ACM international symposium on mixed and augmented reality (pp. 36–45), Tokyo, Japan, October 2003. Google Scholar
- Dunker, J., Hartmann, G., & Stöhr, M. (1996). Single view recognition and pose estimation of 3D objects using sets of prototypical views and spatially tolerant ontour representations. In International conference on pattern recognition (Vol. 4, pp. 14–18), Vienna, Austria, August 1996. Google Scholar
- Ekvall, S., Hoffman, F., & Kragic, D. (2003). Object recognition and pose estimation for robotic manipulation using color coocurrence histograms. In International conference on robots and systems (Vol. 2, pp. 1284–1289), Las Vegas, NV, October 2003. Google Scholar
- Fan, Y., & Balasuriya, A. (2001). Target tracking by underwater robots. In IEEE international conference on systems, man, and cybernetics (pp. 696–701), Tucson, AZ, October 2001. Google Scholar
- Farmer, M., Hsu, R., & Jain, A. (2002). Interacting multiple model (IMM) Kalman filters for robust high speed motion tracking. In International conference on pattern recognition (Vol. 2, pp. 20–23), Québec City, August 2002. Google Scholar
- Fung, J. (2004, November). Parallel computer graphics architectures for computer vision. EyeTap Personal Imaging (ePI) Lab, Edward S. Rogers Dept. of Electrical and Computer Eng., University of Toronto. [Online]. Available: http://www.eyetap.org/about_us/people/fungja/research/.
- Ginhoux, R., & Gutmann, J. (2001). Model-based object tracking using stereo vision. In IEEE international conference on robotics and automation (pp. 1226–1232), Seoul, Korea, May 2001. Google Scholar
- Gong, H., Yang, Q., Pan, C., & Lu, H. (2004). Generalized optical flow in the scale space. In IEEE international conference on image and graphics (pp. 536–539), Hong Kong, China, December 2004. Google Scholar
- Hager, G. D., & Belhumeur, P. N. (1996). Real-time tracking of image regions with changes in geometry and illumination. In IEEE conference on computer vision and pattern recognition (pp. 403–410), San Francisco, CA. Google Scholar
- Han, M., Xu, W., Tao, H., & Gong, Y. (2004). An algorithm for multiple object trajectory tracking. In IEEE conference on computer vision and pattern recognition (pp. 864–871), Washington, DC, June–July 2004. Google Scholar
- Hartley, R., & Kang, S. (2005). Parameter-free radial distortion correction with centre of distortion estimation. In IEEE international conference on computer vision (Vol. 2, pp. 1834–1841), Canberra, Australia, October 2005. Google Scholar
- InforMedia Services. (2006, January). Images. St. Cloud State University. [Online]. Available: http://ims.stcloudstate.edu/handouts/images.htm.
- INTEL. (2004, November). The software vectorization handbook, errata. [Online]. Available: http://www.intel.com/intelpress/vmmx/errata.htm
- Jepson, A., Fleet, D., & El-Maraghi, T. (2001). Robust online appearance models for visual tracking. In IEEE conference on computer vision and pattern recognition (pp. 415–422), Kauai, HI. Google Scholar
- Jia, Z., Balasuriya, A., & Challa, S. (2005). Vision based autonomous vehicles target visual tracking with multiple dynamics models. In IEEE network, sensing and control (pp. 1081–1086), Las Vegas, NV, March 2005. Google Scholar
- Jin, H., Favaro, P., & Soatto, S. (2000). Real-time 3D motion and structure of point-features: a front-end for vision-based control and interaction. In Conference on computer vision and pattern recognition (pp. 778–779), Hilton Head Island, SC. Google Scholar
- Johansson, B., & Moe, A. (2005). Patch-duplets for object recognition and pose estimation. In Canadian conference on computer and robot vision (pp. 9–16), Victoria, Canada, May 2005. Google Scholar
- Jurie, F., & Dhome, M. (2002). Real time robust template matching. In 13th British machine vision conference (pp. 123–132), Cardiff, Wales. Google Scholar
- Kify. (2006, January). Nature wallpapers. [Online]. Available: http://wallpapers.kify.com/nature-wallpapers.htm.
- Kim, S., & Kweon, I. (2003). Robust model-based 3D object recognition by combining feature matching with tracking. In International conference on robotics and automation, (Vol. 2, pp. 2123–2128), Taipei, Taiwan, September 2003. Google Scholar
- Krahnstoever, N., & Sharma, R. (2003). Appearance management and cue fusion for 3D model-based tracking. In Conference on computer vision and pattern recognition (Vol. 2, pp. 249–254), Madison, WI, June 2003. Google Scholar
- Kyrki, V., & Schmock, K. (2005). Integration methods of model-free features for 3D tracking. In Lecture notes in computer science (pp. 557–566). Berlin: Springer. Google Scholar
- Lee, S., Jung, S., & Nevatia, R. (2002). Automatic pose estimation of complex 3D building models. In IEEE workshop on applications of computer vision (pp. 148–152), Orlando, FL, December 2002. Google Scholar
- Lepetit, V., Pilet, J., & Fua, P. (2004). Point matching as a classification problem for fast and robust object pose estimation. In Conference on computer vision and pattern recognition (Vol. 2, pp. 224–250), June 2004. Google Scholar
- Lippiello, V., Siciliano, B., & Villani, L. (2003). Robust visual tracking using a fixed multi-camera system. In IEEE conference on robotics and automation (pp. 3333–3338), Taipei, Taiwan, September 2003. Google Scholar
- Lucas, B., & Kanade, T. (1981). An iterative image registration technique with application to stereo vision. In 7th international joint conference on artificial intelligence (pp. 674–479), Vancouver, Canada, August 1981. Google Scholar
- McKenna, S., Jabri, S., Duric, Z., & Wechsler, H. (2000). Tracking interacting people. In 4th IEEE international conference on automatic face and gesture recognition (pp. 348–353), Grenoble, France, March 2000. Google Scholar
- Ponsa, D., López, A., Serrat, J., Lumbereras, F., & Graf, T. (2005). Multiple vehicle 3D tracking using an unscented Kalman filter. In International IEEE conference on intelligent transportation systems (pp. 1108–1113), Vienna, Austria, September 2005. Google Scholar
- Shreiner, D. (Ed.). (2004). OpenGL reference manual (4th ed.) Boston: Addison-Wesley. Google Scholar
- Virtual New Zealand. (2006, January). Virtual New Zealand photos. [Online]. Available: http://www.virtualoceania.net/newzealand/photos/.
- Webber, J., & Malik, J. (1993). Robust computation of optical flow in a multi-scale differential framework. In 4th international conference on computer vision (pp. 12–20), Berlin, Germany, May 1993. Google Scholar
- Wong, F., Chan, T., Ben Mrad, R., & Benhabib, B. (2004). Mobile-robot guidance in the presence of obstacles. In International conference on flexible automation and intelligent manufacturing (pp. 292–299), Toronto, Canada, July 2004. Google Scholar
- Wu, Y., Hua, G., & Yu, T. (2003). Switching observation models for contour tracking in clutter. In IEEE conference on computer vision and pattern recognition (pp. 295–302), Madison, WI, June 2003. Google Scholar
- Yang, R., Welch, G., & Bishop, G. (2002). Real-time consensus-based scene reconstruction using commodity graphics hardware. In 10th Pacific conference on computer graphics and applications (pp. 225–234), Beijing, China. Google Scholar
- Zhang, Z., Li, J., & Wei, X. (2004). Robust computation of optical flow field with large motion. In IEEE international conference on signal processing (Vol. 1, pp. 893–896), Beijing, China, September 2004. Google Scholar
- Zhao, L., Luo, S., & Liao, L. (2004). 3d object recognition and pose estimation using kernel pca. In 3rd international conference on machine learning and cybernetics (pp. 3258–3262), Shanghai, China, August 2004. Google Scholar