Issues related to PCQM #1
Description
Dear Professor,
We are students with a strong interest in point clouds. While attempting to use the PCQM software, we encountered an issue. We are working with the point cloud file "soldier_vox10_0536.ply," which matches the example file available on GitHub. We decoded this frame's point cloud file using the V-PCC framework, resulting in the decoded "dec.ply" file. When we ran both files through PCQM, the PCQM value returned was only 0.0066999. This result seems to be quite inaccurate. Could you please clarify the expected range for PCQM values under normal circumstances?
In the information file "features_extracted.csv," we noticed that the value of f4 is very small. However, due to its weight being 0.9771, it has a significant impact on the final PCQM result, making it very small. Could there be an anomaly in the f4 value, or is our assumption correct? We would greatly appreciate it if you could find some time to help us address our concerns.
Thank you for your assistance.
Sincerely,
Justin Luzern