TY - GEN
T1 - The synthinel-1 dataset
T2 - 2020 IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2020
AU - Kong, Fanjie
AU - Huang, Bohao
AU - Bradbury, Kyle
AU - Malof, Jordan M.
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/3
Y1 - 2020/3
N2 - Recently deep learning - namely convolutional neural networks (CNNs) - have yielded impressive performance for the task of building segmentation on large overhead (e.g., satellite) imagery benchmarks. However, these benchmark datasets only capture a small fraction of the variability present in real-world overhead imagery, limiting the ability to properly train, or evaluate, models for real-world application. Unfortunately, developing a dataset that captures even a small fraction of real-world variability is typically infeasible due to the cost of imagery, and manual pixel-wise labeling of the imagery. In this work we develop an approach to rapidly and cheaply generate large and diverse synthetic overhead imagery for training segmentation CNNs. Using this approach, we generate and publicly-release a collection of synthetic overhead imagery, termed Synthinel-1, with full pixel-wise building labels. We use several benchmark datasets to demonstrate that Synthinel-1 is consistently beneficial when used to augment real-world training imagery, especially when CNNs are tested on novel geographic locations or conditions.
AB - Recently deep learning - namely convolutional neural networks (CNNs) - have yielded impressive performance for the task of building segmentation on large overhead (e.g., satellite) imagery benchmarks. However, these benchmark datasets only capture a small fraction of the variability present in real-world overhead imagery, limiting the ability to properly train, or evaluate, models for real-world application. Unfortunately, developing a dataset that captures even a small fraction of real-world variability is typically infeasible due to the cost of imagery, and manual pixel-wise labeling of the imagery. In this work we develop an approach to rapidly and cheaply generate large and diverse synthetic overhead imagery for training segmentation CNNs. Using this approach, we generate and publicly-release a collection of synthetic overhead imagery, termed Synthinel-1, with full pixel-wise building labels. We use several benchmark datasets to demonstrate that Synthinel-1 is consistently beneficial when used to augment real-world training imagery, especially when CNNs are tested on novel geographic locations or conditions.
UR - http://www.scopus.com/inward/record.url?scp=85085523132&partnerID=8YFLogxK
U2 - 10.1109/WACV45572.2020.9093339
DO - 10.1109/WACV45572.2020.9093339
M3 - Conference contribution
AN - SCOPUS:85085523132
T3 - Proceedings - 2020 IEEE Winter Conference on Applications of Computer Vision, WACV 2020
SP - 1803
EP - 1812
BT - Proceedings - 2020 IEEE Winter Conference on Applications of Computer Vision, WACV 2020
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 1 March 2020 through 5 March 2020
ER -