Aerial Landing Pad, Unreal Engine Dataset

Dataset of real-world and Unreal Engine captured images of a Landing pad with a KIOS sticker. The airsim dataset contains images from three different backgrounds as well as augmented images generated from these backgrounds. The annotations are available only in YOLO format.

Environment 1 / Augmented : 117 images / 425 images

Environment 2 / Augmented : 163 images / 618 images

Environment 3 / Augmented: 175 images / 647 images

Real video: 422 images

Dataset used in:

Soteriou, C., Kyrkou, C., Kolios, P.S. (2024). Closing the Sim-to-Real Gap: Enhancing Autonomous Precision Landing of UAVs with Detection-Informed Deep Reinforcement Learning. In: Fred, A., Hadjali, A., Gusikhin, O., Sansone, C. (eds) Deep Learning Theory and Applications. DeLTA 2024.

https://link.springer.com/chapter/10.1007/978-3-031-66694-0_11