Large-Scale Fine-Grained Building Classification and Height Estimation for Semantic Urban Reconstruction: Outcome of the 2023 IEEE GRSS Data Fusion Contest

• Published in: IEEE journal of selected topics in applied earth observations and remote sensing Vol. 17; pp. 11194 - 11207
• Main Authors: Liu, Guozhang, Peng, Baochai, Liu, Ting, Zhang, Pan, Yuan, Mengke, Lu, Chaoran, Cao, Ningning, Zhang, Sen, Huang, Simin, Wang, Tao, Lu, Xiaoqiang, Jiao, Licheng, Liu, Qiong, Li, Lingling, Liu, Fang, Liu, Xu, Yang, Yuting, Chen, Kaiqiang, Yan, Zhiyuan, Tang, Deke, Huang, Hai, Schmitt, Michael, Sun, Xian, Vivone, Gemine, Persello, Claudio, Hansch, Ronny
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Buildings; Classification; Convolutional neural networks; data fusion; Data integration; deep learning; Estimation; Feature extraction; fine-grain building classification; Height; Image analysis; Image processing; Lead; Machine learning; monocular height estimation (MHE); Multisensor fusion; Optical imaging; Optical sensors; Radar data; Remote sensing; SAR (radar); Synthetic aperture radar; Teams; transformers
• ISSN: 1939-1404; 2151-1535
• DOI: 10.1109/JSTARS.2024.3403201

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.