An adversarial training framework that uses drone metadata to strip altitude, view-angle, and weather information from detection features, improving UAV object detection by 2 to 4 mAP points on UAVDT and VisDrone2018.
https://www.kdnuggets.com/2018/09/ data-augmentation-bounding-boxes-image-transforms
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Delving into Robust Object Detection from Unmanned Aerial Vehicles: A Deep Nuisance Disentanglement Approach
An adversarial training framework that uses drone metadata to strip altitude, view-angle, and weather information from detection features, improving UAV object detection by 2 to 4 mAP points on UAVDT and VisDrone2018.