Table 6 Linear classification results for different \({\boldsymbol{\lambda }}\) parameters on ModelNet40 and ScanObjectNN datasets after pre-training using PointNet
From: PointMoment: a mixed-moment self-supervised learning approach for 3D Terracotta Warriors
\({\rm{\lambda }}\) | Acc. | |
|---|---|---|
ModelNet40 | ScanObjectNN | |
0.001 | 72.5 | 57.3 |
0.005 | 78.5 | 62.9 |
0.01 | 81.2 | 66.6 |
0.05 | 84.4 | 71.4 |
0.1 | 88.5 | 74.8 |
0.5 | 88.8 | 79.0 |
1 | 88.0 | 76.4 |
5 | 87.8 | 74.1 |
