SemanticSegmentationModel

AP is very lower

11

I use your model to eval. But the map is very low. only 72.12 in hard

Which resolution downloaded from the dl3Dv dataset is used for training?

2

Which resolution downloaded from the DL3DV dataset is used for training?

Is your training configuration set to 20 epochs, and is the total number of data points (len_train) 100,000?

You used a 1 mm threshold instead of the default 20 mm, right, when evaluating with the DTU dataset?