Table 5 PointNet++ approach
From: Object detection for automotive radar point clouds – a comparison
Model part | Implementation details |
|---|---|
Preprocessing | Random up-/down-sampling to 4096 points |
Architecture | Three MSG and FP modules, parameters adopted from [13] |
Clustering | Class-sensitive filtering and clustering, cf. Semantic segmentation network and clustering section |
Class Proposal | Class label via cluster point voting, confidence equal to mean posterior for that class |