ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume 1I-3/W1, 2013
VCM 2013 - The ISPRS Workshop on 3D Virtual City Modeling, 28 May 2013, Regina, Canada
15x15 Augmented WINN
Quasi-exhaustive
Figure 4: Ground truth (GT) and classification results of four different scenes using boosting (row-wise top to bottom: KLOTEN, GRAZ,
VAIHINGEN, WORLDV IEW-2). Streets are displayed black, buildings red, high vegetation s:
it, and low vegetation . KLOTEN
and GRAZ show results of the entire test area without the nDSM channel, VAIHINGEN and WORLDVIEW-2 show zoomed details of
the results using the nDSM channel.
Leung, T. and Malik, J., 2001. Representing and Recognizing
the Visual Appearance of Materials using Three-dimensional
Textons. IJCV 43(1), pp. 29-44.
Martin, D., Fowlkes, C. and Malik, J., 2004. Learning to detect
natural image boundaries using local brightness, color, and tex-
ture cues. IEEE TPAMI 26(5), pp. 530 —549.
Mnih, V. and Hinton, G. E., 2010. Learning to detect roads in
high-resolution aerial images. In: ECCV.
Mnih, V. and Hinton, G. E., 2012. Learning to label aerial images
from noisy data. In: ICML.
Pal, M., 2005. Random forest classifier for remote sensing clas-
sification. International Journal of Remote Sensing 26(1),
pp. 217-222.
Ranzato, M., Huang, E., Boureau, Y. and LeCun, Y., 2007. Un-
supervised Learning of Invariant Feature Hierarchies with Ap-
plications to Object Recognition. In: CVPR.
Rezaei, Y., Mobasheri, M., Zoej, M. V. and Schaepman, M.,
2012. Endmember Extraction Using a Combination of Orthog-
onal Projection and Genetic Algorithm. GRSL 9(2), pp. 161-
165.
Rottensteiner, F.,, Sohn, G., Jung, J., Gerke, M., Baillard, C., Ben-
itez, S. and Breitkopf, U., 2012. The ISPRS benchmark on
urban object classification and 3D building reconstruction. In:
ISPRS Annals, Vol. I-3.
Schapire, R. and Singer, Y., 1999. Improved boosting algorithms
using confidence-rated predictions. Machine Learning 37(3),
pp. 297—336.
Schindler, K., 2012. An Overview and Comparison of Smooth
Labeling Methods for Land-Cover Classification. IEEE TGRS
50(11), pp. 4534-4545.
Schmid, C., 2001. Constructing Models for Content-based Image
Retrieval. In: CVPR.
Schwartz, W., Kembhavi, A., Harwood, D. and Davis, L., 2009.
Human detection using partial least squares analysis. In:
ICCV.
Shao, J. and Foerstner, W., 1994. Gabor wavelets for texture edge
extraction. In: ISPRS Commission III Symposium.
Tokarczyk, P., Montoya, J. and Schindler, K., 2012. An eval-
uation of feature learning methods for high resolution image
classification. ISPRS Annals.
van Coillie, F., Verbeke, L. and Wulf, R. D., 2007. Feature se-
lection by genetic algorithms in object-based classification of
IKONOS imagery for forest mapping in Flanders, Belgium.
Remote Sensing of Environment 110, pp. 476-487.
Viola, P. and Jones, M., 2001. Rapid object detection using a
boosted cascade of simple features. In: CVPR.
Waske, B. and Benediktsson, J., 2007. Fusion of support vector
machines for classification of multisensor data. IEEE TGRS
45(12), pp. 3858-3866.
Winn, A., Criminisi, A. and Minka, T., 2005. Object categoriza-
tion by learned universal visual dictionary. In: ICCV.
Zhu, S., Wu, Y. and Mumford, D., 1997. Minimax Entropy Prin-
ciple and Its Application to Texture Modeling. Neural Compu-
tation 9(8), pp. 1627-1660.
SING]
* GeolC”
® Uni
KEY WOR
ABSTRAC
Tree detect:
model to d
model (CH
constraint c
term which
layouts. We
on three for
There has t
urban and
reconstructi
present in f
and urban f
other appli
received sig
over the pa
detection ar
developed 1
in urban for
of great sig
planning an
2.1 Low-k
Numerous 1
from ALS d
canopy heig
of outer ge
CHM are a:
and crown e
satellite ima
methods de
trees from /
local maxin
(Solberg et
template 1
segmentatio
segmentatio
and multi-sc
Among the
marker-cont
used and 1
