Abstract
An integrated approach to extract depth, efficiently and accurately, from a sequence of images is presented in this paper. The method combines the ability of the stereo processing to acquire highly accurate depth measurements and the efficiency of spatial and temporal gradient analysis. As a result of this integration, depth measurements of high quality are obtained at a speed approximately ten times greater than that of stereo processing. Without any a priori information of the locations of the points in the scene, the correspondence problem in stereo processing is computationally expensive. In our approach, we use spatial and temporal gradient (STG) analysis, which has been shown to provide depth with great efficiency, but limited accuracy, to guide the matching process of stereo. The camera motion used in the approach can be either lateral or axial. Extensive experiments on real scenes have shown the ability of the integrated approach to acquire depth with a mean error of less than 3%.
Similar content being viewed by others
References
Besl PJ (1988) Active, optical range imaging sensors. Machine Vision Appl 1:127–152
Mitiche A (1988) Motion understanding: robot and human vision, pp 81–99. Kluwer Academic, Boston, MA
Grosso E, Sandini G, Tistarelli M (1989) 3-D Object reconstruction using stereo and motion. IEEE Trans Syst Man Cybernetics 19:1465–1476
Symosek PF, Bhanu B, Snyder S, Roberts B (1990) Motion and binocular stereo for passive ranging. Proceedings, Image Understanding Workshop, Pittsburgh, PA, September 11–13, 1990, Morgan Kaufman, San Mateo, CA, pp 358–363
Vernon D, Tistarelli M (1990) Using camera motion to estimate range for robotic parts manipulation. IEEE Trans Robotics Automation 6:509–521
Sridhar B, Suorsa R (1990) Integration of motion and stereo sensors in passive ranging systems. The American Control Conference
Horn BKP (1986) Robot vision. MIT Press, Cambridge, MA
Dhond UR, Aggarwal JK (1989) Structure from stereo — a review. IEEE Trans Syst Man Cybernetics 19:1489–1510
Nevatia R (1976) Depth measurement by motion stereo. Comput Graph Image Processing 5:203–214
Bolles RC, Baker HH, Hannah MJ (1993) The JISCT stereo evaluation. Proceedings, [ARPA] Image Understanding Workshop, Washington, DC, April 18–21, 1993, Morgan Kaufman, San Mateo, CA, pp 263–274
Skifstad K, Jain R (1989) Range estimation from intensity gradient analysis. Machine Vision Appl 2:81–102
Horn BKP, Schunk BG (1981) Determining optical flow. Artif Intell 17:185–203
Nguyen TC, Huang TS (1992) Quantization errors in axial motion stereo on rectangular-tessellated image sensors. Proceedings of the International Conference on Pattern Recognition, pp 13–16
Press WH, Flannery BP, Teukolsky SA, Vetterling WI (1988) Numerical recipes in C. Cambridge University Press, Cambridge
Carmona C, Ayache A, Krey C (1990) A study of an axial stereovision system. Proceedings of International Conference on Automation, Robotics and Computer Vision, Singapore, pp 1101–1106
Marapane S, Trivedi MM (1994) Multi-primitive hierarchical (MPH) stereo analysis. IEEE Trans Patt Anal Machine Intell 16:227–240
Aggarwal JK, Davis LS, Martin WN (1981) Correspondence processes in dynamic scene analysis. Proceedings of the IEEE 69:562–572
Dalmia AK, Trivedi MM (1993) A real time implementation of structure estimation using image streams. Proceedings of the SPIE — Sensor Fusion VI 2059:248–259
Dalmia AK, Trivedi MM (1995) High-speed extraction of 3D structures using image streams obtained by a single camera. Computer Vision, Graphics and Image Processing: Image Understanding, 1994. To appear 1995
Dalmia AK, Trivedi MM (1994) On integrating depth from motion and stereo. Proceedings of the IEEE International Conference on Systems, Man, and Cybernetics, San Antonio, Tex, pp 61–66
Dalmia AK, Trivedi MM (1994) Acquisition of 3D structure of selectable quality from image streams. Proceedings of the IEEE Workshop on Applications of Computer Vision, Sarasota, Fla., pp 289–296
Dalmia AK, Trivedi MM (1994) Active depth extraction using image streams. Proceedings of the SPIE — Sensor Fusion and Aerospace Application II, Orlando, Fla.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Dalmia, A.K., Trivedi, M. Depth extraction using a single moving camera: an integration of depth from motion and depth from stereo. Machine Vis. Apps. 9, 43–55 (1996). https://doi.org/10.1007/BF01214359
Issue Date:
DOI: https://doi.org/10.1007/BF01214359