) by z keying.
a real image
ge IS(i,j) and
rdinates. The
ly created by
proper rang-
real scene to
| with coordi-
e two depth
e that has the
je 10(i,j). The
n any desired
virtual world
Figure 9, part
virtual object
object (e.g., a
1 wall.
bjects in real
ich extraction
orresponding
eshold. Also,
prior to or in
ISd(i,j)
ying
5.2 Real-Time Z keying with The Stereo Machine
Figure 10 shows an example sequence of the demonstra-
tion. In this demonstration a real person walks around in a
synthetic room, and correct relationships with virtual
objects in the room are achieved. The demonstration can
perform z keying at the rate of 15 frames/sec.
Figure 10: Example images demonstrate z keying
with the CMU video-rate stereo machine
6. Conclusion
This paper has presented the CMU video-rate stereo
machine and its application to virtual reality. The machine
is capable of producing a dense 200 x 200 depth map,
aligned with intensity information, at 30 frames per sec-
ond. This performance represents a one or two order of
magnitude improvement over the current state of the art in
415
passive stereo range mapping. Such a capability opens up
a new class of applications of 3D vision, and we have pre-
sented z-keying in the area of visual media interaction.
7. References
Faugeras, O., 1992. What can be seen in three dimen-
sions with an uncalibrated stereo rig?, In Computer Vision
- ECCV ‘92, pp 563-578.
Faugeras, O., et al., 1993. Real time correlation based ste-
reo: algorithm, implementations and applications,
Research Report 2013, INRIA Sophia-Antipolis.
Kanade, T., et al., 1995. Video-Rate Z Keying: A New
Method for Merging Images, CMU-RI-TR-95-38.
Kimura, et al., 1995. CMU Video-Rate stereo machine, In
Proc. of Mobile Mapping Symposium 95, Columbus, OH-
USA, May 24-26.
Matthies, L. H., 1992. Stereo vision for planetary rovers:
stochastic modeling to near real time implementation.
International Journal of Computer Vision, 8 (1), pp 71-91.
Nakahara, T. and Kanade, T., 1992. Experiments in multi-
ple-baseline stereo. Technical report, Carnegie Mellon
University, Computer Science Department, August.
Nishihara, H.K., 1990. Real-time implementation of a sign-
correlation algorithm for image-matching. (Draft) Teleos
Research, February.
Okutomi, M. and Kanade, T., 1993. A multi-baseline ste-
reo. In Proc. of Computer Vision and Pattern Recognition
1993.
Okutomi, M., et al, 1992. A multiple-baseline stereo
method. In Proc. of DARPA Image Understanding Work-
shop, pp. 409-426.
Tsai, R.Y., 1987. A versatile camera calibration technique
for high-accuracy 3D machine vision metrology using off-
the-shelf TV cameras and lenses, IEEE Journal of Robot-
ics and Automation, Vol.RA-3, No.4.
Webb, J., 1993. Implementation and performance of fast
parallel multi-baseline stereo vision. In Proc. of Image
Understanding Workshop, pp.1005-1012.
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B5. Vienna 1996
