TY - GEN
T1 - Learning class- and location-specific priors for urban semantic labeling with CNNs
AU - Kellenberger, Benjamin
AU - Volpi, Michele
AU - Tuia, Devis
PY - 2017/5/10
Y1 - 2017/5/10
N2 - This paper addresses the problem of semantic image labeling of urban remote sensing images into land cover maps. We exploit the prior knowledge that cities are composed of comparable spatial arrangements of urban objects, such as buildings. To do so, we cluster OpenStreetMap (OSM) building footprints into groups with similar local statistics, corresponding to different types of urban zones. We use the per-cluster expected building fraction to correct for over- and underrepresentations of classes predicted by a Convolutional Neural Network (CNN), using a Conditional Random Field (CRF). Results indicate a substantial improvement in both numerical and visual accuracy of the labeled maps.
AB - This paper addresses the problem of semantic image labeling of urban remote sensing images into land cover maps. We exploit the prior knowledge that cities are composed of comparable spatial arrangements of urban objects, such as buildings. To do so, we cluster OpenStreetMap (OSM) building footprints into groups with similar local statistics, corresponding to different types of urban zones. We use the per-cluster expected building fraction to correct for over- and underrepresentations of classes predicted by a Convolutional Neural Network (CNN), using a Conditional Random Field (CRF). Results indicate a substantial improvement in both numerical and visual accuracy of the labeled maps.
U2 - 10.1109/JURSE.2017.7924537
DO - 10.1109/JURSE.2017.7924537
M3 - Conference paper
AN - SCOPUS:85020233104
SN - 9781509058082
T3 - 2017 Joint Urban Remote Sensing Event, JURSE 2017
BT - 2017 Joint Urban Remote Sensing Event, JURSE 2017
PB - Institute of Electrical and Electronics Engineers Inc.
CY - Dubai
T2 - 2017 Joint Urban Remote Sensing Event, JURSE 2017
Y2 - 6 March 2017 through 8 March 2017
ER -