Underwater Marker-Based Pose-Estimation With Associated Uncertainty

Petter Risholm, Peter Ørnulf Ivarsen, Karl Henrik Haugholt, Ahmed Mohammed; Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops, 2021, pp. 3713-3721

We propose a system for 6-DoF estimation of Aruco markers with associated uncertainties in the challenging underwater environment. A state-of-the-art object detection framework (EfficientDet) was adapted to predict the corner locations of Aruco markers, while dropout sampling at inference time is used to estimate the predictive 6-DoF pose uncertainty. A dataset of Aruco markers captured in a wide variety of turbidities, with ground truth position of the corner locations, was gathered and used to train the network to robustly predict the 6-DoF pose. We report median translational errors of 2.6cm at low turbidity (8.5m attenuation length) and up to 10.5cm at high turbidities (0.3m attenuation length). The respective uncertainty, reported as interquartile ranges (IQRs), range from 3.2cm up to 27.9cm. The rotational median errors varied from 5.6 (deg) to 10.7 (deg) with IQRs of 6.4 (deg) to 26.2 (deg). We also discuss how the pose uncertainty can be applied to reduce the risk in a subsea intervention operation.