When the option “rectify to deep map” is selected during the collection of the deep learning dataset, what changes will occur in the color image in this case? Can the corrected color image be used for training deep learning models?
-
When triggering the option, if there are defects in the point cloud, the color image will exclude the reflective parts.
-
As 2D images may lose some pixels, which is not ideal for training, it is better not to use them for deep learning.