search

pqhieu / jsis3d

[CVPR'19] JSIS3D: Joint Semantic-Instance Segmentation of 3D Point Clouds
https://pqhieu.com/research/cvpr19/
MIT License
177 stars 36 forks source link

How to prepare my own data set #12

Closed zhangmenghan closed 5 years ago

zhangmenghan commented 5 years ago

In your code, you say follow the scripts folder to learn how to prepare the data set.But there are four files in the folder, what steps should be followed to prepare. Also, how do I set the num_points parameter when I do the prepare_h5.py file: batch = room_to_blocks(fname, **num_points,** size=1.0, stride=0.5) I would also like to ask which of the following data sets is appropriate for processing Stanford3dDataset_v1.2_Aligned_Version or Stanford3dDataset_v1.2

pqhieu commented 5 years ago

Assume that you use the aligned version, you only need to run collect_annotations.py and then prepare_h5.py to prepare the data, the others are standalone scripts to calculate some dataset statistics. num_points is set to 4096 in my experiments.

zhangmenghan commented 5 years ago

Thanks a lot, I will try it now to prepare my dataset