Closed iamcoming233 closed 1 year ago
Is it necessary to calibrate the image pair before it is fed into the stereo matching network?
In practical use, it is necessary to calibrate the image pair and then input the trained network to predict the disparity. Since the right image in the training process is generated according to the left image and depth map (disparity map), the training input image pair can be considered to have been calibrated. As a result, it is necessary to calibrate the image pair to acquire disparity
Is it necessary to calibrate the image pair before it is fed into the stereo matching network?