High-Definition 5MP Stereo Vision Sensing for Robotics
Leaf Jiang, Matthew Holzel, Bernhard Kaplan, Hsiou-Yuan Liu, Sabyasachi Paul, Karen Rankin, Piotr Swierczynski
TL;DR
This study tackles the calibration challenge in high-resolution stereo vision by introducing an online auto-calibration method and a dual stereo-matcher framework (Hammerhead for real-time, GroundTruth for offline high-accuracy). It derives and validates a depth-uncertainty scaling law showing that depth precision improves with the square root of the pixel count, and demonstrates that 5MP stereo can achieve dense, metric-depth point clouds only when calibration is sufficiently accurate. The GroundTruth-based evaluation provides a robust, real-world measure of depth quality beyond traditional planar targets. Practically, the work shows HD stereo cameras with online calibration can extend reliable perception to longer ranges, delivering higher-density 3D maps for robotics and automation.
Abstract
High-resolution (5MP+) stereo vision systems are essential for advancing robotic capabilities, enabling operation over longer ranges and generating significantly denser and accurate 3D point clouds. However, realizing the full potential of high-angular-resolution sensors requires a commensurately higher level of calibration accuracy and faster processing -- requirements often unmet by conventional methods. This study addresses that critical gap by processing 5MP camera imagery using a novel, advanced frame-to-frame calibration and stereo matching methodology designed to achieve both high accuracy and speed. Furthermore, we introduce a new approach to evaluate real-time performance by comparing real-time disparity maps with ground-truth disparity maps derived from more computationally intensive stereo matching algorithms. Crucially, the research demonstrates that high-pixel-count cameras yield high-quality point clouds only through the implementation of high-accuracy calibration.