Stereo Matching

Stereoscopic matching is one of the core technologies in computer vision, aiming to recover the three-dimensional structure of the real world from two-dimensional images. Its primary goal is to calculate the disparity for each pixel in the reference image, which refers to the horizontal displacement of corresponding pixels between the left and right images. This technology has significant application value in fields such as autonomous driving, augmented reality, and robotic navigation.