TY - GEN
T1 - Adaptive Bins for Monocular Height Estimation from Single Remote Sensing Images
AU - Chen, Sining
AU - Shi, Yilei
AU - Xiong, Zhitong
AU - Zhu, Xiao Xiang
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - Monocular height estimation is of great importance in generating 3D city models from single remote sensing images, while it is a challenging task due to the ill-posed nature of the problem. To address the issue, we propose to adopt adaptive bins (AdaBins) for the network design, which enhances the representation capability of the network with the classification-regression paradigm and the incorporation of local features and global context via a vision transformer encoder. Besides, to weaken the biases of the trained networks caused by the long-tailed nature of the dataset, a head-tail cut is conducted for different treatments of head and tail pixels. Experiments show that improvements are expected with the proposed network on the proposed GBH dataset.
AB - Monocular height estimation is of great importance in generating 3D city models from single remote sensing images, while it is a challenging task due to the ill-posed nature of the problem. To address the issue, we propose to adopt adaptive bins (AdaBins) for the network design, which enhances the representation capability of the network with the classification-regression paradigm and the incorporation of local features and global context via a vision transformer encoder. Besides, to weaken the biases of the trained networks caused by the long-tailed nature of the dataset, a head-tail cut is conducted for different treatments of head and tail pixels. Experiments show that improvements are expected with the proposed network on the proposed GBH dataset.
KW - adaptive bins
KW - hybrid regression
KW - monocular height estimation
KW - vision transformer
UR - http://www.scopus.com/inward/record.url?scp=85178349332&partnerID=8YFLogxK
U2 - 10.1109/IGARSS52108.2023.10281953
DO - 10.1109/IGARSS52108.2023.10281953
M3 - Conference contribution
AN - SCOPUS:85178349332
T3 - International Geoscience and Remote Sensing Symposium (IGARSS)
SP - 7015
EP - 7018
BT - IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2023 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2023
Y2 - 16 July 2023 through 21 July 2023
ER -