Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces TESO, a novel method for online tracking of the essential matrix in stereo camera systems using stochastic optimization on the essential manifold. TESO employs a robust kernel correlation-based loss function over tentative correspondences and adaptive online stochastic optimization, achieving low CPU and memory requirements without requiring data-driven training. Experiments on MAN TruckScenes and KITTI datasets demonstrate TESO's ability to track rotational calibration drift with high precision, revealing systematic inconsistencies and improving depth accuracy significantly when correcting reference calibrations.
Achieve 20x improvement in rotation precision and 50x improvement in depth accuracy for stereo camera calibration by simply correcting reference calibration errors using the new TESO method.
Maintaining long-term accuracy of stereo camera calibration parameters is important for autonomous systems' perception. This work proposes Online Tracking of Essential Matrix by Stochastic Optimization (TESO). The core mechanisms of TESO are: 1) a robust loss function based on kernel correlation over tentative correspondences, 2) an adaptive online stochastic optimization on the essential manifold. TESO has low CPU and memory requirements, relies on a few hyperparameters, and eliminates the need for data-driven training, enabling the usage in resource-constrained online perception systems. We evaluated the influence of TESO on geometric precision, rectification quality, and stereo depth consistency. On the large-scale MAN TruckScenes dataset, TESO tracks rotational calibration drift with 0.12 deg precision in the Y-axis (critical for stereo accuracy) while the X- and Z-axes are five times more precise. Tracking applied to sequences with simulated drift shows similar precision with respect to the reference as tracking applied to no-drift sequences, indicating the tracker is unbiased. On the KITTI dataset, TESO revealed systematic inconsistencies in extrinsic parameters across stereo pairs, confirming previous published findings. We verified that intrinsic decalibration affected these errors, as evidenced by the conflicting behavior of the rectification and depth metrics. After correcting the reference calibration, TESO improved its rotation precision around the Y-axis 20 times to 0.025 deg and its depth accuracy 50 times. Despite its lightweight design, direct optimization of the proposed TESO loss function alone achieves accuracy comparable to that of neural network-based single-frame methods.