Unlocking Drone Perception in Low AGL Heights: Progressive Semi-Supervised Learning for Ground-to-Aerial Perception Knowledge Transfer
- 15 May 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 26 (7) , 9516-9528
Abstract
We explore the novel challenge of drone perception across varying low AGL (above ground level) heights, a task essential for dynamic tasks, unlike the fixed ground viewpoint in autonomous driving. Supervised learning for this incurs high annotation costs, and current semi-supervised methods struggle with viewpoint differences. In this paper, we introduce ground-to-aerial perception knowledge transfer and propose a progressive semi-supervised learning framework for drone perception using only labeled data from the ground viewpoint and unlabeled data from flying viewpoints. The framework hinges on four key components: 1) a dense viewpoint sampling strategy, segmenting the vertical flight height range into evenly distributed intervals; 2) nearest neighbor pseudo-labeling, inferring labels of the nearest neighbor viewpoint using a model learned on the preceding viewpoint; 3) MixView, generating augmented images among different viewpoints to mitigate viewpoint differences; and 4) a progressive distillation strategy, gradually learning until reaching the maximum flying height. To validate our approach, we create both synthesized and real-world datasets. Extensive experimental analyses reveal a remarkable relative accuracy improvement of 25.7% and 16.9% for the synthesized dataset and the real world, respectively. Code and datasets are available on https://github.com/FreeformRobotics/Progressive-Self-Distillation-for-Ground-to-Aerial-Perception-Knowledge-Transfer.Keywords
Funding Information
- Shenzhen Science and Technology Program (RCBS20231211090736065)
- Basic and Applied Basic Research Foundation of Guangdong Province (2023A151511)
- Guangdong Natural Science Fund (2024A1515010252)
- Longgang District Shenzhen’s “Ten Action Plan” for Supporting Innovation Projects (LGKCSDPT2024002/LGKCSDPT2024003)
This publication has 49 references indexed in Scilit:
- Dense depth distillation with out-of-distribution simulated imagesKnowledge-Based Systems, 2023
- Boosting LightWeight Depth Estimation via Knowledge Distillation2023
- Deep learning-based object detection in low-altitude UAV datasets: A surveyImage and Vision Computing, 2020
- UAVid: A semantic segmentation dataset for UAV imageryISPRS Journal of Photogrammetry and Remote Sensing, 2020
- A survey on semi-supervised learningMachine Learning, 2019
- VisDrone-DET2018: The Vision Meets Drone Object Detection in Image Challenge Results2019
- Deep Neural Network for Autonomous UAV Navigation in Indoor Corridor EnvironmentsProcedia Computer Science, 2018
- AirSim: High-Fidelity Visual and Physical Simulation for Autonomous Vehicles2017
- Structure-from-Motion Revisited2016
- The Cityscapes Dataset for Semantic Urban Scene Understanding2016