Pointnet2.PyTorch icon indicating copy to clipboard operation
Pointnet2.PyTorch copied to clipboard

Using code for just xyz attributes

Open narges-tk opened this issue 2 years ago • 2 comments

Hi! Many thanks for sharing your code. Could you please let me know how I can run your repository for point clouds that have just xyz attributes? Thanks in advance. Best regards, Narges

narges-tk avatar Dec 01 '23 12:12 narges-tk

@narges-tk hello, have you solved the issue? it's bothering me now

13717630148 avatar Oct 12 '24 07:10 13717630148

@narges-tk hello, have you solved the issue? it's bothering me now

Hi! Yes! In train_custom_cls.py, I defined ninputfeatures parameter as bellow: parser.add_argument('--ninputfeatures', type=int, default=6, help='Number of initial(input) features in point cloud') For a point cloud with XYZ, ninputfeatures=3. Then use it as below: Model = Models[args.model] model = Model(args.ninputfeatures, args.nclasses) This is what I remember. Hope that helps!

narges-tk avatar Oct 14 '24 21:10 narges-tk