TY - JOUR
T1 - Large-scale fine-grained building classification and height estimation for semantic urban reconstruction
T2 - Outcome of the 2023 IEEE GRSS Data Fusion Contest
AU - Liu, Guozhang
AU - Peng, Baochai
AU - Liu, Ting
AU - Zhang, Pan
AU - Yuan, Mengke
AU - Lu, Chaoran
AU - Cao, Ningning
AU - Zhang, Sen
AU - Huang, Simin
AU - Wang, Tao
AU - Lu, Xiaoqiang
AU - Jiao, Licheng
AU - Liu, Qiong
AU - Li, Lingling
AU - Liu, Fang
AU - Liu, Xu
AU - Yang, Yuting
AU - Chen, Kaiqiang
AU - Yan, Zhiyuan
AU - Tang, Deke
AU - Huang, Hai
AU - Schmitt, Michael
AU - Sun, Xian
AU - Vivone, Gemine
AU - Persello, Claudio
AU - Hansch, Ronny
N1 - Publisher Copyright:
© 2008-2012 IEEE.
PY - 2024
Y1 - 2024
N2 - This article presents the scientific outcomes of the 2023 Data Fusion Contest (DFC23) organized by the Image Analysis and Data Fusion Technical Committee of the IEEE Geoscience and Remote Sensing Society. The contest consists of two tracks investigating the fusion of optical and synthetic aperture radar data for: 1) fine-grained roof type classification and 2) height estimation. During the development phase, 1000 people registered for the contest, while at the end 55 and 35 teams competed during the test phase in the two tracks, respectively. This article presents the methods and results obtained by the first and second-ranked teams of each track. In Track 1, both winning teams leveraged pretraining, modern network architectures, model ensembles, and measures to cope with the imbalanced class distribution. The solutions to Track 2 are more diverse and are characterized by modern multitask learning approaches. The data of this contest is openly available to the community for further research, development, and refinement of machine learning methods.
AB - This article presents the scientific outcomes of the 2023 Data Fusion Contest (DFC23) organized by the Image Analysis and Data Fusion Technical Committee of the IEEE Geoscience and Remote Sensing Society. The contest consists of two tracks investigating the fusion of optical and synthetic aperture radar data for: 1) fine-grained roof type classification and 2) height estimation. During the development phase, 1000 people registered for the contest, while at the end 55 and 35 teams competed during the test phase in the two tracks, respectively. This article presents the methods and results obtained by the first and second-ranked teams of each track. In Track 1, both winning teams leveraged pretraining, modern network architectures, model ensembles, and measures to cope with the imbalanced class distribution. The solutions to Track 2 are more diverse and are characterized by modern multitask learning approaches. The data of this contest is openly available to the community for further research, development, and refinement of machine learning methods.
KW - Convolutional neural networks
KW - data fusion
KW - deep learning
KW - fine-grain building classification
KW - monocular height estimation (MHE)
KW - transformers
KW - ITC-GOLD
KW - ITC-ISI-JOURNAL-ARTICLE
U2 - 10.1109/JSTARS.2024.3403201
DO - 10.1109/JSTARS.2024.3403201
M3 - Article
AN - SCOPUS:85194035594
SN - 1939-1404
VL - 17
SP - 11194
EP - 11207
JO - IEEE Journal of selected topics in applied earth observations and remote sensing
JF - IEEE Journal of selected topics in applied earth observations and remote sensing
ER -