search

Sangminhong / ACL-SPC_PyTorch

Official implementation of the paper "ACL-SPC: Adaptive Closed-Loop system for Self-Supervised Point Cloud Completion" (CVPR 2023)
34 stars 5 forks source link

About PCN Dataset #10

Open ljj621 opened 8 months ago

ljj621 commented 8 months ago

Thank you very much for your excellent work. I noticed in the supplementary materials of your paper that you trained and evaluated on PCN dataset. I have a few questions:

  1. How do you train? Should we train on each category or all categories? I tried training directly on all categories, but the effect was not as good as expected
  2. How is evaluation on unseen categories implemented? Very much looking forward to your answer.