fit_surface_principal_axes — 3D geometry op

• Data kinds: points → table

• Call: import fullseye as fs; fs.ledger.fit_surface_principal_axes(points, degree: 'int' = 2, *, isotropy_ratio: 'float' = 0.95) -> 'dict' (to call the implementation directly, import measure3d; measure3d.fit_surface_principal_axes(points, degree: 'int' = 2, *, isotropy_ratio: 'float' = 0.95) -> 'dict'; from the registry, ops3d.get("fit_surface_principal_axes"))

• GPU: none (CPU reference implementation only)

Usage

Rotate a slanted warped surface to PCA axes, then fit polynomial z'=f(x',y').

The two largest principal axes span the local x/y plane and the smallest is local z. Returns center, row-vector axes, monomial powers, coefficients, local and original-frame fitted_points, unit normal at the centroid, and gradient-normalised geometric rms. degree is the total polynomial degree. A cloud whose two largest variances differ by less than isotropy_ratio is rejected because its in-plane principal directions are not identifiable. For degree one the returned normal agrees with fit_plane3 up to sign.

References (sample data, literature)

• Sample-data catalog (download URLs / licences) — 2-D uses skimage.data (BSD/public domain) plus synthetic images; 3-D lists download URLs for real data sources (Stanford, PDS, …).

• Operator provenance and references — the sources of the research/methods this op family came from.

Runnable examples (verified samples that actually call this op)

• primitive_fitting_3d — py -3.11 examples_3d/primitive_fitting_3d.py

Ops the type connects to (they accept table as input)

fuse_to_voxel · mesh_select_lod

Same category (geometry)

line_from_2points · plane_from_3points · angle_3points · angle_between_lines · angle_between_planes · angle_line_plane · distance_point_plane · distance_point_line


*Provenance: measure3d.py — 3D operator registry. This per-op note is generated by tools/opdocs.py md (do not hand-edit).*

© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.