Abstract | ||
---|---|---|
Automatically finding good and general remote sensing representations allows to perform transfer learning on a wide range of applications - improving the accuracy and reducing the required number of training samples. This paper investigates development of generic remote sensing representations, and explores which characteristics are important for a dataset to be a good source for representation learning. For this analysis, five diverse remote sensing datasets are selected and used for both, disjoint upstream representation learning and downstream model training and evaluation. A common evaluation protocol is used to establish baselines for these datasets that achieve state-of-the-art performance. As the results indicate, especially with a low number of available training samples a significant performance enhancement can be observed when including additionally in-domain data in comparison to training models from scratch or fine-tuning only on ImageNet (up to 11% and 40%, respectively, at 100 training samples). All datasets and pretrained representation models are published online. |
Year | DOI | Venue |
---|---|---|
2020 | 10.1109/IGARSS39084.2020.9324501 | IGARSS 2020 - 2020 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM |
Keywords | DocType | Citations |
Remote sensing, representation learning, transfer learning, convolutional neural networks | Conference | 0 |
PageRank | References | Authors |
0.34 | 0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Maxim Neumann | 1 | 0 | 0.68 |
André Susano Pinto | 2 | 0 | 1.69 |
Xiaohua Zhai | 3 | 0 | 0.34 |
Neil Houlsby | 4 | 0 | 0.34 |