typed op• Data kinds: points → volume
• Call: fullseye.apply(img, "tb_plane_segmentation", a=0.5, b=0.5) (the 2-D model is one image plus two scalar knobs a,b∈[0,1])
*No figure: this op takes points as input. A Studio program starting from an image cannot reach that type — see the runnable examples below for how it is used.*
Iteratively extracts up to max_planes planes via repeated RANSAC (remaining residual points are -1).
> The detailed description below is the original text — the summary and the headings are translated.
残り点集合に :func:ransac_fit.ransac_plane を掛け、その最大 consensus 平面の
inlier 数が `min_inliers` 以上なら新ラベルを与えて除去 → 残りで再検出、を繰り返す。
複数の床/壁/階段状の面を一度に分離する(単一平面適合の pcseg との差)。inlier が
`min_inliers` に満たなくなった時点で停止し、以降の点は残差 -1(球や複雑物体はここに残る)。
Args:
points: (N,3) 点群。
thresh: 点-平面距離の inlier しきい値(距離、要 > 0)。
min_inliers: 平面として採用する最小 inlier 数(要 >= 3)。
max_planes: 抽出する平面の最大枚数(要 >= 1)。
iters: 各 RANSAC 反復数。
seed: 乱数シード(決定論。各平面で seed+平面index を使う)。
Returns:
labels: (N,) int。検出順(=consensus 大きい順に近い)に 0,1,2,... を平面へ付与、
どの平面にも属さない残差点は -1。空入力は shape (0,)。
2-D 進化レジストリへ橋渡しした 3d の op `plane_segmentation。実装は同じで、呼び出し規約だけ op(v, a, b) に合わせてある。a が max_planes(既定 5)、b が iters`(既定 300)を振る。
• 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.
• (none yet)
volume as input)identity · vol_gaussian · vol_median · vol_erode · vol_dilate · vol_threshold · vol_reg_dilate · vol_reg_erode
typed)tb_points_to_voxel · tb_estimate_point_normals · tb_iss_keypoints · tb_angle_3points · tb_project_points · tb_render_point_depth · tb_statistical_outlier_removal · tb_radius_outlier_removal
*Provenance: ops.py — 2D 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.