Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection (FPFH)

Reported on 2 benchmarks across 1 task · 1 paper · 2 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology2 results

Anomaly DetectiononMVTEC 3D-AD
Detection AUROC· 2022-03-10
0.782
best: 93.7 (AST)
SOTA
Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection arXiv:2203.05550
Anomaly DetectiononMVTEC 3D-AD
Segmentation AUROC· 2022-03-10
0.978
best: 97.6 (AST)
SOTA
Back to the Feature: Classical 3D Features are (Almost) All You Need for 3D Anomaly Detection arXiv:2203.05550