Hi， I want to create my own dataset.
when i visualize track2_validation/input/Pair001.npy, the normalized_depth's max value is not equal 1,
For my own images，If I use a monocular depth estimation algorithm to predict the depth map of an image, how can I converting the predicted depth map into the "ref_center_dis" and "normalized_depth"?
Hi， I want to create my own dataset. when i visualize track2_validation/input/Pair001.npy, the normalized_depth's max value is not equal 1,
For my own images，If I use a monocular depth estimation algorithm to predict the depth map of an image, how can I converting the predicted depth map into the "ref_center_dis" and "normalized_depth"?