How to write test results as ply

[alien19]

Hi, I was lucky to try your code and test it on the S3DIS dataset, also I noticed the presence of "write_ply_color" and "write_ply_rgb" but it is not obvious how to be used in the test code so, if there a way to do so, it would be appreciated. specifically, how can points argument be passed to the functions

[X-Lai]

You can use the following function to output the .obj file, and open it with MeshLab.

def write_obj(points, colors, out_filename):
    N = points.shape[0]
    fout = open(out_filename, 'w')
    for i in range(N):
        c = colors[i]
        fout.write('v %f %f %f %d %d %d\n' % (points[i, 0], points[i, 1], points[i, 2], c[0], c[1], c[2]))
    fout.close()

[alien19]

thanks @X-Lai , I have checked this function too in a previous issue, but I can't understand what is the points object that is to be passed to the function.

[X-Lai]

points is a numpy array object (np.ndarray) of the shape (N, 3), which means xyz of all points.

[alien19]

@X-Lai I understand this thanks. but when I run the test.py script the output .npy files which contain pred and labels are 1d arrays that's why I can't figure out from where points object would be constructed!

[X-Lai]

The .npy files produced by test.py don't contain the xyz position information, you may need to use the corresponding input file with the same filename for visualization.

[alien19]

I managed to do the visualization successfully thanks @X-Lai , but I'm wondering about the correct order of the classes so as to list them correctly in the s3dis_names.txt file since there is a mismatch between what is present in the test_log file (downloaded with the model) and the colors dictionary in the code

[X-Lai]

Thank you for pointing out this issue, I find that the order of s3dis_names.txt is problematic. Please refer to the https://github.com/yanx27/Pointnet_Pointnet2_pytorch for the correct class mapping.