Preprint
Aug 2026
Scalix: Uncertainty-Aware Scale-Consistent Monocular SLAM
Scalix is proposed, a real-time monocular SLAM framework that achieves metric-scale state estimation by integrating learned depth cues into a probabilistic factor-graph formulation, leading to improved scale consistency through multi-view data associations.
Sebastián Barbas Laina, Tian-Yi Zhang, Panagiotis Petropoulakis et al.
· 0 citations