Scannet data processing while testing MonoSDF

[saulgooodman]

Hello, after reading the content about custom data in the .md file, I would like to confirm how to obtain all the data:

1.Firstly, download Scannet from the official website and preprocess it according to the official tutorial. 2.Then, use scannet_to_monosdf.py to obtain cameras.npz. 3.Use extract_monocular_cues.py on the Scannet preprocessed with the Scannet official tutorial. 4.Finally, all the data is obtained, including RGB (from original scannet dataset), depth & normal (from extract_monocular_cues.py), cameras.py(from scannet_to_monosdf.py).

I'm not sure if my understanding is correct because I noticed that both scannet_to_monosdf.py and extract_monocular_cues.py generate depth maps, and the depth maps generated by the two are not the same.

Thank you!