Bridging the Domain Gap in Satellite Pose Estimation: a Self-Training Approach based on Geometrical Constraints
This work addresses the annotation cost problem for researchers and practitioners in satellite pose estimation, but it is incremental as it builds on existing self-training and geometrical constraint methods.
The paper tackles the problem of unsupervised domain adaptation for satellite pose estimation by proposing a self-training framework that uses domain-agnostic geometrical constraints and fine-grained segmentation to reduce annotation costs. It achieved first place in a competition, demonstrating effective adaptation to target domains.
Recently, unsupervised domain adaptation in satellite pose estimation has gained increasing attention, aiming at alleviating the annotation cost for training deep models. To this end, we propose a self-training framework based on the domain-agnostic geometrical constraints. Specifically, we train a neural network to predict the 2D keypoints of a satellite and then use PnP to estimate the pose. The poses of target samples are regarded as latent variables to formulate the task as a minimization problem. Furthermore, we leverage fine-grained segmentation to tackle the information loss issue caused by abstracting the satellite as sparse keypoints. Finally, we iteratively solve the minimization problem in two steps: pseudo-label generation and network training. Experimental results show that our method adapts well to the target domain. Moreover, our method won the 1st place on the sunlamp task of the second international Satellite Pose Estimation Competition.