English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

UAVs have become an essential photogrammetric measurement as they areaffordable, easily accessible and versatile. Aerial images captured from UAVshave applications in small and large scale texture mapping, 3D modelling,object detection tasks, DTM and DSM generation etc. Photogrammetric techniquesare routinely used for 3D reconstruction from UAV images where multiple imagesof the same scene are acquired. Developments in computer vision and deeplearning techniques have made Single Image Depth Estimation (SIDE) a field ofintense research. Using SIDE techniques on UAV images can overcome the need formultiple images for 3D reconstruction. This paper aims to estimate depth from asingle UAV aerial image using deep learning. We follow a self-supervisedlearning approach, Self-Supervised Monocular Depth Estimation (SMDE), whichdoes not need ground truth depth or any extra information other than images forlearning to estimate depth. Monocular video frames are used for training thedeep learning model which learns depth and pose information jointly through twodifferent networks, one each for depth and pose. The predicted depth and poseare used to reconstruct one image from the viewpoint of another image utilisingthe temporal information from videos. We propose a novel architecture with two2D CNN encoders and a 3D CNN decoder for extracting information fromconsecutive temporal frames. A contrastive loss term is introduced forimproving the quality of image generation. Our experiments are carried out onthe public UAVid video dataset. The experimental results demonstrate that ourmodel outperforms the state-of-the-art methods in estimating the depths.

Self-supervised monocular depth estimation from oblique UAV videos