Keypoint-Guided Efficient Pose Estimation and Domain Adaptation for Micro Aerial Vehicles

Visual detection of micro aerial vehicles (MAVs) is an important problem in many tasks such as vision-based swarming of MAVs. This paper studies vision-based 6D pose estimation to detect a 3D bounding box of a target MAV and then estimate its 3D position and 3D attitude. The 3D attitude information is critical to better estimate the target's velocity since the attitude and motion are dynamically coupled. In this paper, we propose a novel 6D pose estimation method, whose novelties are threefold. First, we propose a novel centroid point-guided keypoint localization network that outperforms the state-of-the-art methods in terms of both accuracy and efficiency. Second, while there are no publicly available real-world datasets for 6D pose estimation for MAVs up to now, we propose a high-quality dataset based on an automatic dataset collection method. Third, since the dataset is collected in an indoor environment but detection tasks are usually in outdoor environments, we propose a self-training-based unsupervised domain adaption method to transfer the method from indoor to outdoor. Finally, we show that the estimated 6D pose especially the 3D attitude can significantly help improve the target's velocity estimation.