Table 1: The average and the standard deviation of estima-
tion errors over 130 steps.
before after
modiﬁcation modiﬁcation
average [cm] 0.268 0.0394
standard
deviation [cm] 0.102 0.00739
anteed. With this modiﬁcation, errors involved in the
parameters before the modiﬁcation are drastically re-
duced and, therefore, signiﬁcant improvement of ac-
curacy in estimation for a long trajectory is realized,
with keeping the method incremental.
The proposed method focuses on the ﬁxation point
for evaluating geometrical consistency, however, us-
ing other feature points together may improve estima-
tion accuracy further. An adaptive application of the
bundle adjustment, i,e., effectively selecting steps to
which the local bundle adjustment is applied, will be
promising to reduce computational cost. These inves-
tigations are our future direction.
ACKNOWLEDGEMENTS
This work is in part supported by Grant-in-Aid for
Scientiﬁc Research of the Ministry of Education, Cul-
ture, Sports, Science and Technology of Japan under
the contract of 13224051 and 18049046.
REFERENCES
H. Aoki, B. Schiele and A. Pentland (2000): Realtime Per-
sonal Positioning System for Wearable Computers, Vi-
sion and Modeling Technical Report, TR-520, Media
Lab. MIT.
B. Clarkson, K. Mase and A. Pentland (2000): Recognizing
User’s Context from Wearable Sensors: Baseline Sys-
tem, Vision and Modeling Technical Report, Vismod
TR-519, Media Lab. MIT.
A. J. Davison, W. W. Mayol and D. W. Murray (2003):
Real-Time Localization and Mapping with Wearable
Active Vision, Proc. of ISMAR, pp. 18–27.
A. J. Davison and D. W. Murray (1998): Mobile Robot Lo-
calization using Active Vision, Proc. of ECCV, Vol. 2,
pp. 809–825.
G. N. DeSouza and A. C. Kak (2002): Vision for Mobile
Robot Navigation: A Survey, IEEE Trans. on PAMI,
Vol. 24, No. 2, pp. 237–267.
M. W. M. G. Dissanayake, P. Newman, S. Clark, H. F.
Durrant-Whyte and M. Csorba (2001): A Solution
to the Simultaneous Localization and Map Building
(SLAM) Problem, IEEE Trans. on RA, Vol. 17, No. 3,
pp. 229–241.
N. Gonc¸alves and H. Ara
´
ujo (2002): Estimation of 3D Mo-
tion from Stereo Images, Proc. of ICPR, Vol. I, pp.
335–338.
J. E. Guivant and E. Nebot (2001): Optimization of the
Simultaneous Localization and Map-Building Algo-
rithm for Real-Time Implementation, IEEE Trans. on
RA, Vol. 17, No. 3, pp. 242–257.
R. Hartley and A. Zisserman (2000): Multiple View Geom-
etry in Computer Vision, Cambridge Univ. Press.
B. D. Lucas and T. Kanade (1981): An Iterative Image Reg-
istration Technique with an Application to Stereo Vi-
sion, Proc. of IJCAI, pp. 674–679.
N. Molton and M. Brady (2003): Practical Structure and
Motion from Stereo When motion is Unconstrained,
Int. J. of Computer Vision, Vol. 39, No. 1, pp. 5–23.
T. Nakagawa, T. Okatani and K. Deguchi (2004): Active
Construction of 3D Map in Robot by Combining Mo-
tion and Perceived Images, Proc. of ACCV, vol. 1, pp.
563–568.
A. Sugimoto, W. Nagatomo and T. Matsuyama (2004): Es-
timating Ego Motion by Fixation Control of Mounted
Active Cameras, Proc. of ACCV, pp. 67–72.
A. Sugimoto and T. Ikeda (2004): Diverging Viewing-Lines
in Binocular Vision: A Method for Estimating Ego
Motion by Mounted Active Cameras, Proc. of the 5th
Workshop on Omnidirectional Vision, Camera Net-
works and Non-classical Cameras, pp. 67–78.
Y. Sumi, Y. Ishiyama and F. Tomita (2004): 3D Localiza-
tion of Moving Free-Form Objects in Cluttered Envi-
ronments, Proc. of ACCV, vol. 1, pp. 43-48.
S. Thrun (2002): Robotic Mapping: A Survey, Exploring
Artiﬁcial Intelligence in the New Millennium, Morgan
Kaufmann.
C. Tomasi and T. Kanade (1991): Detection and Tracking
of Point Features, CMU Technical Report, CMU-CS-
91-132.
B. Triggs, P. McLauchlan, R. Hartley and A. Fitzgibbon
(2000): Bundle Adjustment – A Modern Synthesis,
Vision Algorithms: Theory and Practice (B. Triggs, A.
Zisserman and R. Szeliski eds.) LNCS, Vol. 1883, pp.
298–372, Springer.
B. L. Welch (1938): The Signiﬁcance of the Difference be-
tween Two Means when the Population Variances are
Unequal, Biometrika, Vol. 29, pp. 350–362.
M. Werman, S. Banerjee, S. Dutta Roy and M. Qiu (1999):
Robot Localization Using Uncalibrated Camera In-
variants, Proc. of CVPR, Vol. II, pp. 353–359.
Z. Zhang and Y. Shan (2001): Incremental Motion Estima-
tion through Local Bundle Adjustment, Technical Re-
port MSR-TR-01-54, Microsoft Research.
Z. Zhang and Y. Shan (2003): Incremental Motion Esti-
mation through Modiﬁed Bundle Adjustment, Proc.
ICIP, Vol.II, pp.343–346.
REDUCING ACCUMULATED ERRORS IN EGO-MOTION ESTIMATION USING LOCAL BUNDLE ADJUSTMENT
425