A Neural Tension Operator for Curve Subdivision across Constant Curvature Geometries
For researchers in geometric modeling and computer graphics, this work provides a data-driven approach to adaptively control subdivision tension across constant curvature geometries, though it is incremental as it applies existing neural prediction to a known problem.
The paper introduces a neural tension predictor that replaces the global tension parameter in interpolatory subdivision schemes with per-edge insertion angles, enabling a single network to operate across Euclidean, spherical, and hyperbolic geometries. On 240 held-out curves, the learned predictor achieves lower bending energy and angular roughness than fixed-tension baselines, and on an out-of-distribution ISS example, bending energy drops by 41% and angular roughness by 68% with a modest increase in Hausdorff distance.
Interpolatory subdivision schemes generate smooth curves from piecewise-linear control polygons by repeatedly inserting new vertices. Classical schemes rely on a single global tension parameter and typically require separate formulations in Euclidean, spherical, and hyperbolic geometries. We introduce a shared learned tension predictor that replaces the global parameter with per-edge insertion angles predicted by a single 140K-parameter network. The network takes local intrinsic features and a trainable geometry embedding as input, and the predicted angles drive geometry-specific insertion operators across all three spaces without architectural modification. A constrained sigmoid output head enforces a structural safety bound, guaranteeing that every inserted vertex lies within a valid angular range for any finite weight configuration. Three theoretical results accompany the method: a structural guarantee of tangent-safe insertions; a heuristic motivation for per-edge adaptivity; and a conditional convergence certificate for continuously differentiable limit curves, subject to an explicit Lipschitz constraint verified post hoc. On 240 held-out validation curves, the learned predictor occupies a distinct position on the fidelity--smoothness Pareto frontier, achieving markedly lower bending energy and angular roughness than all fixed-tension and manifold-lift baselines. Riemannian manifold lifts retain a pointwise-fidelity advantage, which this study quantifies directly. On the out-of-distribution ISS orbital ground-track example, bending energy falls by 41% and angular roughness by 68% with only a modest increase in Hausdorff distance, suggesting that the predictor generalises beyond its synthetic training distribution.