FoV-Net: Rotation-Invariant CAD B-rep Learning via Field-of-View Ray Casting

TL;DR

FoV-Net introduces a rotation-invariant B-rep learning framework for 3D CAD analysis, capturing local and global context through ray casting and graph attention, significantly improving robustness to rotations.

Contribution

It is the first B-rep learning method that achieves rotation invariance by combining local surface geometry with global context via ray casting and graph attention.

Findings

Achieves state-of-the-art results on B-rep classification and segmentation.

Demonstrates robustness to arbitrary rotations.

Requires less training data for strong performance.

Abstract

Learning directly from boundary representations (B-reps) has significantly advanced 3D CAD analysis. However, state-of-the-art B-rep learning methods rely on absolute coordinates and normals to encode global context, making them highly sensitive to rotations. Our experiments reveal that models achieving over 95% accuracy on aligned benchmarks can collapse to as low as 10% under arbitrary $\mathbf{SO}(3)$ rotations. To address this, we introduce FoV-Net, the first B-rep learning framework that captures both local surface geometry and global structural context in a rotation-invariant manner. Each face is represented by a Local Reference Frame (LRF) UV-grid that encodes its local surface geometry, and by Field-of-View (FoV) grids that capture the surrounding 3D context by casting rays and recording intersections with neighboring faces. Lightweight CNNs extract per-face features, which are propagated over the B-rep graph using a graph attention network. FoV-Net achieves state-of-the-art performance on B-rep classification and segmentation benchmarks, demonstrating robustness to arbitrary rotations while also requiring less training data to achieve strong results.