TY - JOUR
T1 - Filtering Specialized Change in a Few-Shot Setting
AU - Hermann, Martin
AU - Saha, Sudipan
AU - Zhu, Xiao Xiang
N1 - Publisher Copyright:
© 2008-2012 IEEE.
PY - 2023
Y1 - 2023
N2 - The aim of change detection in remote sensing usually is not to find all differences between the observations, but rather only specific types of change, such as urban development, deforestation, or even more specialized categories like roadwork. However, often there are no large public datasets available for very fine-grained tasks, and to collect the amount of training data needed for most supervised learning methods is very costly and often prohibitive. For this reason, we formulate the problem of few-shot filtering, where we are provided with a relatively large change detection dataset and, at test time, a few instances of one particular change type that we try to 'filter out' of the learned changes. For example, we might train on data of general urban change, and, given some samples of building construction, aim to only predict instances of these on the test set, all without any explicit labels for buildings in the training data. We further investigate a fine-Tuning approach to this problem and assess its performance on a public dataset that we adapt to be used in this novel setting.
AB - The aim of change detection in remote sensing usually is not to find all differences between the observations, but rather only specific types of change, such as urban development, deforestation, or even more specialized categories like roadwork. However, often there are no large public datasets available for very fine-grained tasks, and to collect the amount of training data needed for most supervised learning methods is very costly and often prohibitive. For this reason, we formulate the problem of few-shot filtering, where we are provided with a relatively large change detection dataset and, at test time, a few instances of one particular change type that we try to 'filter out' of the learned changes. For example, we might train on data of general urban change, and, given some samples of building construction, aim to only predict instances of these on the test set, all without any explicit labels for buildings in the training data. We further investigate a fine-Tuning approach to this problem and assess its performance on a public dataset that we adapt to be used in this novel setting.
KW - Change detection
KW - deep learning
KW - few-shot filtering
KW - few-shot learning
UR - http://www.scopus.com/inward/record.url?scp=85147312013&partnerID=8YFLogxK
U2 - 10.1109/JSTARS.2022.3231915
DO - 10.1109/JSTARS.2022.3231915
M3 - Article
AN - SCOPUS:85147312013
SN - 1939-1404
VL - 16
SP - 1185
EP - 1196
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 -