Can a model be trained on "generated" GT?

[zheng zhenkai]

Hi, I would like to ask about the feasibility of training a model from the code provided using generated dataset. I have a set of RGB images obtained from several open-source data and I have passed the dataset into the model provided in https://github.com/compphoto/BoostingMonocularDepth to produce ground truth. If I were to use this "generated" dataset, what should be the depth scale and maximum depth I should input? I notice that there is a fixed depth scale and maximum depth for NYU and KITTI datasets in the code such as
self.depth_scale = 256.0 in line 24 of datasets_list.py

[Minsoo Song]

I think a model can be trained on the generated dataset. However, you have to determine the criteria for the depth minimum and maximum values covering all samples in the dataset. In the case of a dataset based on a hardware sensor (KITTI, NYU..), max and min value of the depth map is determined by this sensor. However, when using the generated dataset, I think the user should adjust the sample value so that all samples in the dataset have the same depth range.

[zheng zhenkai]

thanks for the clarification!