Question about PP-score in case of single traversal

[baurst]

Hi,

thank you for your great work, and thank you for publishing the code! I have a question regarding your baseline MODEST-PP (R0), which you mention in your paper (the one that does not require multiple traversals).

In the paper you write: "The seed labels are constructed by the exact same process as described in section 3, except we replace the edge weights in Equation 4 by spatial proximity: [...] and do not perform any PP-score-based filtering on the clusters generated by DBSCAN"

Does that mean, that PP score is not used at all in this case? Because if we only have a single traversal, I believe we will divide by zero in formula (3) (log(1) = 0).

This leads me to believe, that it is not used at all (and the fact that it is not used here, when affinity_type == '3d_l2_distance' )

Does this mean that for MODEST-PP, you are basically doing DBSCAN for each point cloud individually, but the edge weights are just the nearest neighbor distances?

Thank you very much for your help!

[YurongYou]

Hi Stefan,

Thanks for your interest! I think you understanding is correct. As stated in the paper, for MODEST-PP, we are not using PP at all in the whole process. When doing the clustering in this case, the edge weights are just the nearest neighbor distances.

[baurst]

That makes sense, thank you very much for the quick response! So for the self improvement you just replace the old seed labels with the network predictions, without any filtering (or confidence thresholding), correct?

[YurongYou]

Yes, you are correct!