International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXIX-B3, 2012
XXII ISPRS Congress, 25 August - 01 September 2012, Melbourne, Australia
--- plc ——p2b «9 p2b cp2v
150% i
(a) CT, AUGN (b) CT, Salt-and-Pepper
50096 -— 15.0%
450% i 14.0%
i 1306 -
400% ©
35096 : 1106 3s
300% 10.0%
(e) RT, AWGN (f) RT, Salt-and-Pepper
0 10 20 30 4 0 © 7 8 % 100 0 10 20 30 40 SO 60 70 8 90 100
0 10 20 30 4 0 © 7 8 % 100 0 10 20 30 4) 9 D D D D 100
0 10 20 30 40 SO 60 70 8 90 100 0 310 20 30 40 50 & 70 80 90 10
(c) CT, Shadowed (d) CT, Gamma
0 10 20 30 40 SO 60 70 8 90 10 0 10 20 30 40 SO 60 70 80 90 100
(g) RT, Shadowed (h) RT, Gamma
Figure 7: Errors in unoccluded areas for all penalty functions on degenerated input images employing census and rank transform.
k
L
(k) P»;, Baseline
(a) Ground Truth (b) AWGN (c) Salt-and-Pepper (d) Shadowed (e) Gamma
(f) P», Baseline (g) P1, AWGN (h) P», Salt-and-Pepper
(1) Pj, AWGN (m) P»;, Salt-and-Pepper
(i) P51, Shadowed () P51, Gamma
(n) P»;, Shadowed (0) P»;, Gamma
Figure 8: Degenerated input images (row one) and corresponding disparity maps obtained with P»; (row two) and P»; (row three) using
the census transform.
REFERENCES
Banz, C., Blume, H. and Pirsch, P., 2011a. Real-time semi-global
matching disparity estimation on the gpu. In: Computer Vision
Workshops (ICCV Workshops), 2011 IEEE International Confer-
ence on, pp. 514 —521.
Banz, C., Hesselbarth, S., Flatt, H., Blume, H. and Pirsch, P,
2011b. Real-time stereo vision system using semi-global match-
ing disparity estimation: Architecture and fpga-implementation.
Transactions on High-Performance Embedded Architectures and
Compilers, Springer.
Ess, A., Leibe, B. and Van Gool, L., 2007. Depth and appearance
for mobile scene analysis. In: IEEE Intl. Conf. Computer Vision,
pp. 1 -8.
Haller, I. and Nedevschi, S., 2012. Design of interpolation func-
tions for subpixel-accuracy stereo-vision systems. Image Pro-
cessing, IEEE Transactions on 21(2), pp. 889 —898.
Hirschmüller, H., 2008. Stereo Processing by Semiglobal Match-
ing and Mutual Information. IEEE Trans. Pattern Analysis and
Machine Intelligence 30(2), pp. 328-341.
Hirschmiiller, H. and Scharstein, D., 2009. Evaluation of
stereo matching costs on images with radiometric differences.
IEEE Trans. Pattern Analysis and Machine Intelligence 31(9),
pp. 1582-1599.
Scharstein, D. and Szeliski, R., 2002. A taxonomy and evalua-
tion of dense two-frame stereo correspondence algorithms. Intl.
Journal of Computer Vision 47(1), pp. 7-42.
Scharstein, D. and Szeliski, R., 2012. The Middlebury Stereo
Pages. http://vision.middlebury.edu/stereo/.
Zabih, R. and Woodfill, J., 1994. Non-parametric local trans-
forms for computing visual correspondence. In: European Con-
ference on Computer Vision, pp. 151-158.
KE
AB
The
Stat
rela
the
val
The
orie
det:
the
the
Fin:
sup
The
of 1
Spa
pos
app
aut
me:
dist
bet
In «
on
nec
the:
or
sys
exti
deti
thei
met
sys
din
reg
ima
The
the
ran
lea
doc