fit op• Data kinds: keypoints → matrix
• Call: import fullseye as fs; fs.ledger.hom_vector_to_proj_hom_mat2d(src, dst) -> 'np.ndarray' (to call the implementation directly, import fit_transform; fit_transform.hom_vector_to_proj_hom_mat2d(src, dst) -> 'np.ndarray'; from the registry, opscalib.get("hom_vector_to_proj_hom_mat2d"))
• HALCON equivalent: hom_vector_to_proj_hom_mat2d (the HALCON reference is a useful guide to its meaning and parameters)
• GPU: none (CPU reference implementation only)
Finds a 3x3 projective transform (homography, DLT) from 4 or more correspondences (hom_vector_to_proj_hom_mat2d).
> The detailed description below is the original text — the summary and the headings are translated.
本モジュールの契約どおり点は (row, col): 返る H は (row, col, 1) の同次ベクトルに左から掛けて (row', col', w) を与える(transforms.projective_trans_point_2d(H, row, col) と整合)。数値安定化のため両点集合を Hartley 正規化してから DLT を解き H = T_dst⁻¹ · H_n · T_src で戻す。
(旧実装は DLT 行を (col, row) 順で組んでいたため、(row, col, 1) に適用すると 2 座標を取り違えた行列を返していた — 2026-09-02 実測 max err 27 px。)
• camera_calibration family guide
• 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.
• The canonical algorithm (author, year) and its uses are named in the family usage guide above.
• poc_panorama_drift — py -3.11 examples/poc_panorama_drift.py
• poc_wound_area_tracking — py -3.11 examples/poc_wound_area_tracking.py
matrix as input)fit)vector_to_rigid · vector_to_similarity · vector_angle_to_rigid
*Provenance: fit_transform.py — CALIB 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.