Training an inference on alternate dataset

[abhigoku10]

@Yang7879 thanks for open sourcing the wonderfull work , i had few queries
Q1 have you trained the architecture on the available other dataset like semanttic Kitti and 3D dataset
Q2 If not trained can we follow the same training pipeline , if trained can you please share the pre-trained model
Q3 can we use the currently pre-trained model to test on custom dataset which less number of point cloud density

Thanks in advance

[Bo Yang]

Hi @abhigoku10 , for the time being, we haven't trained the model on other datasets yet, but I believe it is able to be trained if you follow a similar data preprocessing strategy. For your own dataset where the number of points are fewer, you could upsample it to be a same number (e.g., 4k or 2k maybe).

[abhigoku10]

@Yang7879 can you name some upsmapling methods which i can use to increase the density of point cloud , i generally do aggregation but besides this anything else would be helpful

[Bo Yang]

@abhigoku10 , the simplest approach is to randomly duplicate some points for upsampling.