Open Yipingggg opened 1 year ago
In the code, seems all the input data are normalized, but they are simulated data. When training linemod or custom dataset, how to deal with the input correspondence? What normalization should I do, or I don't need to normalize? Thanks!
In the code, seems all the input data are normalized, but they are simulated data. When training linemod or custom dataset, how to deal with the input correspondence? What normalization should I do, or I don't need to normalize? Thanks!