yanx27 / GACNet

Pytorch implementation of 'Graph Attention Convolution for Point Cloud Segmentation'
166 stars 31 forks source link

Question about the data prepare functions in model.py #6

Open limt15 opened 5 years ago

limt15 commented 5 years ago

Thank you very much for sharing the fantastic code! I wondered about the data format of the functionsquare_distance in the model.py script when I was reading it. Could you please tell more details of the input data below:

Input: src: source points, [B, N, C] dst: target points, [B, M, C]

What‘s the [B,N,C] and [B,M,C] looks like? What’s the difference between the [B,M,C] and [B, npoint, C]? And what‘s the means about these capital letters?