Does it generalize to unseen in-the-wild point clouds?
coreqode opened this issue · comments
Hi, this is amazing work.
Does the pre-trained model generalise to in-the-wild point clouds? or it is specific to trained datasets only?
It generalizes well to real-world kinect style scanner based point clouds.
Is there any codebase for testing the pre-trained models in in-the-wild scenarios?
This can be achieved by changing the dataset using your custom sensor data, then run the testing.