Persistent Homology Features for Robust 3D Point Cloud Classification

N. Meters, S. Hoffmann

We introduce a pipeline that extracts persistent homology descriptors from 3D point clouds and integrates them as auxiliary features into standard classification architectures. The topological features capture global shape properties invariant to noise, occlusion, and sampling density. On ModelNet40 and ScanObjectNN, augmenting PointNet++ with our descriptors improves robustness to 60% point dropout by 8.3 percentage points with negligible computational overhead.