Event

Proposing a monocular depth-assisted two-stage LiDAR-camera calibration optimization method

1 reports · 1 sources

Researchers proposed a monocular depth-assisted two-stage joint optimization calibration method for LiDAR-camera extrinsic parameters, replacing traditional grayscale-intensity mutual information constraints with depth constraints to improve calibration accuracy and robustness under complex lighting and material conditions. In the coarse calibration stage, a composite calibration board combining ArUco markers and a circular hole array is used; an SVD-optimized ICP algorithm solves for initial extrinsic parameter values without requiring preset initial values. In the fine calibration stage, the DepthPro monocular depth estimation network is introduced to generate dense predicted depth maps, and a joint optimization objective function is constructed combining geometric constraints and normalized mutual information distance (NID). For testing, a Livox Mid-360 hybrid semi-solid-state LiDAR was selected, and sensor data were collected in multiple scenarios across 7 groups of different observation poses; the calibration ground truth was constructed via a manual-assisted calibration procedure.