EgoHDM is an innovative online egocentric-inertial motion capture system that provides near real-time localization and dense scene mapping, enhancing human motion estimation in various terrains.
OmniColor colorizes point clouds with a 360-degree camera, optimizing camera poses for accurate image-to-geometry mapping without feature extraction, resulting in improved 3D reconstruction in robotic navigation and scene reconstruction tasks.
In this work, we study the challenging problem of how to effectively update an existing fingerprint database given a batch of unlabelled (crowdsourced) signals that may consist of new APs.