SkyScenes: A Synthetic Dataset for Aerial Scene Understanding

Sahil S Khose*, Anisha Pal, Aayushi Agarwal, . Deepanshi, Judy Hoffman, Prithvijit Chattopadhyay ;

"Real-world aerial scene understanding is limited by a lack of datasets that contain densely annotated images curated under a diverse set of conditions. Due to inherent challenges in obtaining such images in controlled real-world settings, we present , a synthetic dataset of densely annotated aerial images captured from Unmanned Aerial Vehicle (UAV) perspectives. We carefully curate images from to comprehensively capture diversity across layouts (urban and rural maps), weather conditions, times of day, pitch angles and altitudes with corresponding semantic, instance and depth annotations. Through our experiments using , we show that (1) models trained on generalize well to different real-world scenarios, (2) augmenting training on real images with data can improve real-world performance, (3) controlled variations in can offer insights into how models respond to changes in viewpoint conditions (height and pitch), weather and time of day, and (4) incorporating additional sensor modalities (depth) can improve aerial scene understanding. Our dataset and associated generation code are publicly available at: https://hoffman-group.github.io/SkyScenes/ Aerial scene understanding, Synthetic-to-Real generalization, Segmentation, Domain Generalization, Synthetic Data"