Computing points.npy, view_id.npy, pairs.txt
Opened this issue · 6 comments
Dear authors,
Thank you for your contribution. In testing your code, I assume that the cameras.npz files follow the format from NeuS and IDR. However, you also load the sparse pointcloud in the points.npy and view_id.npy. Could you please provide data samples and/or explain the structure of these files?
Futhermore you use a pairs.txt file which I assume is a precomputed set of close source views for each reference views. How is this computed?
Hi, @sergiobdrl, thanks for attention! We have released your mentioned files. To obtain the pairs.txt, you can modify this file to compute the source views for each reference view.
Thank your for your answers!
Hi, have you find out how to get the points.npy and view_id.npy files? Thanks very much.
Hi, @sergiobdrl, thanks for attention! We have released your mentioned files. To obtain the pairs.txt, you can modify this file to compute the source views for each reference view.
Hi, Thank you for your contribution. I want to train my own data, could you explain how to get the points.npy and view_id.npy files? Thanks very much.
In the file that you mention, https://github.com/GhiXu/ACMP/blob/master/colmap2mvsnet_acm.py#L412, you preselect 20 views. I'm curious to know how this hyperparameter affects the results...
@sergiobdrl Dear authors, I'm confused too. Could you please share the code to estimate the point cloud?
Your colmap results look much better than the NeuS.
Yours:
Neus:
