.png){kind=icon}
.png){kind=icon}
Return
THE Benchmark: Transferable Representation Learning for Monocular Height Estimation
.png){kind=icon}
OA
AI
DOI：10.1109/TGRS.2023.3311764.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Generating 3-D city models rapidly is crucial for many applications. Monocular height estimation (MHE) is one of the most efficient and timely ways to obtain large-scale geometric information. However, existing works focus primarily on training and testing models using unbiased datasets, which does not align well with real-world applications. Therefore, we propose a new benchmark dataset to study the transferability of height estimation models in a cross-dataset setting. To this end, we first design and construct a large-scale benchmark dataset for cross-dataset transfer learning on the height estimation task. This benchmark dataset includes a newly proposed large-scale synthetic dataset, a newly collected real-world dataset, and four existing datasets from different cities. Next, a new experimental protocol, few-shot cross-dataset transfer, is designed. Furthermore, in this article, we propose a scale-deformable convolution (SDC) module to enhance the window-based Transformer for handling the scale-variation problem in the height estimation task. Experimental results have demonstrated the effectiveness of the proposed methods in traditional and cross-dataset transfer settings.
Keywords:
Benchmark
cross-dataset transfer
remote sensing
synthetic data
transfer learning
Transformer
Journal
IF:
8.6
Papers:
2.1W
Citations:
10.7W