PiCoCo: Pixelwise Contrast and Consistency Learning for Semisupervised Building Footprint Segmentation
![Thumbnail](/xmlui/bitstream/handle/10234/196138/PiCoCo_2021_IEEE.pdf.jpg?sequence=4&isAllowed=y)
Visualitza/
Impacte
![Google Scholar](/xmlui/themes/Mirage2/images/uji/logo_google.png)
![Microsoft Academico](/xmlui/themes/Mirage2/images/uji/logo_microsoft.png)
Metadades
Mostra el registre complet de l'elementcomunitat-uji-handle:10234/9
comunitat-uji-handle2:10234/43662
comunitat-uji-handle3:10234/43643
comunitat-uji-handle4:
INVESTIGACIONMetadades
Títol
PiCoCo: Pixelwise Contrast and Consistency Learning for Semisupervised Building Footprint SegmentationData de publicació
2021-10-11Editor
IEEEISSN
1939-1404; 2151-1535Cita bibliogràfica
J. Kang, Z. Wang, R. Zhu, X. Sun, R. Fernandez-Beltran and A. Plaza, "PiCoCo: Pixelwise Contrast and Consistency Learning for Semisupervised Building Footprint Segmentation," in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 14, pp. 10548-10559, 2021, doi: 10.1109/JSTARS.2021.3119286.Tipus de document
info:eu-repo/semantics/articleVersió
info:eu-repo/semantics/publishedVersionParaules clau / Matèries
Resum
Building footprint segmentation from high-resolution
remote sensing (RS) images plays a vital role in urban planning, disaster response, and population density estimation. Convolutional
neural networks (CNNs) have ... [+]
Building footprint segmentation from high-resolution
remote sensing (RS) images plays a vital role in urban planning, disaster response, and population density estimation. Convolutional
neural networks (CNNs) have been recently used as a workhorse for
effectively generating building footprints. However, to completely
exploit the prediction power of CNNs, large-scale pixel-level annotations are required. Most state-of-the-art methods based on CNNs
are focused on the design of network architectures for improving
the predictions of building footprints with full annotations, while
few works have been done on building footprint segmentation with
limited annotations. In this article, we propose a novel semisupervised learning method for building footprint segmentation, which
can effectively predict building footprints based on the network
trained with few annotations (e.g., only 0.0324 km2 out of 2.25-km2
area is labeled). The proposed method is based on investigating
the contrast between the building and background pixels in latent
space and the consistency of predictions obtained from the CNN
models when the input RS images are perturbed. Thus, we term the
proposed semisupervised learning framework of building footprint segmentation as PiCoCo, which is based on the enforcement of
Pixelwise Contrast and Consistency during the learning phase. Our
experiments, conducted on two benchmark building segmentation
datasets, validate the effectiveness of our proposed framework as
compared to several state-of-the-art building footprint extraction
and semisupervised semantic segmentation methods. [-]
Publicat a
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol. 14 (2021)Entitat finançadora
Jiangsu Province Science Foundation for Youths | National Natural Science Foundation of China | Jiangsu Higher Education Institution | Ministerio de Ciencia, Innovación y Universidades (España) | APRISA | Generalitat Valenciana | FEDER-Junta de Extremadura | European Union
Codi del projecte o subvenció
BK20210707 | 62101371 | 62076241 | RTI2018-098651-B-C54 | PID2019- 110315RB-I00 | GV/2020/167 | GR18060 | info:eu-repo/grantAgreement/EC/H2020/734541
Drets d'accés
info:eu-repo/semantics/openAccess
Apareix a les col.leccions
- INIT_Articles [752]