Renbiao Jin

KLTNet: Learning Sparse Feature Tracking for Robust and Accurate Monocular Visual-Inertial Odometry

KLTNet is proposed, a lightweight learning-based, plug-and-play sparse feature tracker designed to replace classical KLT trackers in KLT-based VIO front ends and predicts anisotropic confidence weights supervised through differentiable multi-view triangulation, which can be used as observation weights in compatible VIO estimators.