dataset
rsy6318 opened this issue · comments
In the paper, 20 meshes of PU dataset were used to train, but in the code, there are 40 meshes. Can it influence the result?
I have retrain the network and get worse results.
Did you train the model for a long enough time, or did the training converge?
The scale of the point clouds affects the metric values. The same normalization should be applied to get comparable results.
Hi @luost26, I would like to clarify for the pretrained model (and its results mentioned in the paper), was 20 meshes sampled or was it 40? If only 20 were sampled, is it possible to provide the list of meshes?
Thank you!
Hi @luost26, I would like to clarify for the pretrained model (and its results mentioned in the paper), was 20 meshes sampled or was it 40? If only 20 were sampled, is it possible to provide the list of meshes?
Thank you!
The training set size is 40. 20 is a mistake in the paper. Sorry about that.
@luost26, fantastic, that's all good then. Great work and thanks for the clarification!