scene_flow3d op• Data kinds: points × points → pose
• Call: import fullseye as fs; fs.ledger.rigid_flow(pts0, pts1, max_iter: 'int' = 20) -> 'dict' (to call the implementation directly, import scene_flow3d; scene_flow3d.rigid_flow(pts0, pts1, max_iter: 'int' = 20) -> 'dict'; from the registry, ops3d.get("rigid_flow"))
• Return value through the ledger: fullseye.ledger.rigid_flow(...) returns **only the declared out type pose** (the underlying function also returns auxiliary values). When you need what was dropped, use fullseye.ledger.rigid_flow.raw(...) or call scene_flow3d.rigid_flow directly.
Estimate the single rigid motion that explains pts0 -> pts1, via nearest-neighbour correspondences and Kabsch (ICP-like).
> The detailed description below is the original text — the summary and the headings are translated.
恒等変換から出発し、毎反復で現姿勢の点群から `pts1` への最近傍対応を取り、
:func:registration.kabsch で閉形式に (R, t) を求めて累積する。対応 index が
前反復と一致(= 収束)するか `max_iter` で停止する。収束判定は index の
安定性のみに依存するためスケール不変(絶対 epsilon を使わない)。
Args:
pts0: (N, 3) 時刻 0 の点群(N >= 3、回転を一意に決めるため)。
pts1: (M, 3) 時刻 1 の点群(M >= 1、N と一致不要)。
max_iter: ICP 反復上限。
Returns:
dict: ``{"R": (3,3) 回転, "t": (3,) 並進, "rmse": 整合後の点-最近傍
RMS 距離}``。R は真の回転(det=+1)。rmse は実測値(詐称なし)。
Raises:
ValueError: 形状不正、pts0 < 3 点、または pts1 が空。
• 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.
• scene_flow_rigid — py -3.11 examples_3d/scene_flow_rigid.py
pose as input)fuse_to_voxel · pose_error · bundle_adjust · mean_reprojection_error · optimize_pose_graph · relative_pose · mean_edge_error · rotation_translation_error
scene_flow3d)nearest_neighbor_flow · smooth_flow
*Provenance: scene_flow3d.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.