transform op• Data kinds: voxel → images
• Call: import fullseye as fs; fs.ledger.voxel_to_mips(vol) (to call the implementation directly, import match3d; match3d.voxel_to_mips(vol); from the registry, ops3d.get("voxel_to_mips"))
3-D → maximum-intensity projections along the three orthogonal directions. The entry point for applying 2-D methods (e.g. accel's 2-D NCC).
> The detailed description below is the original text — the summary and the headings are translated.
入力 `(D,H,W) に対し [max(axis=0), max(axis=1), max(axis=2)]` の list を返す。形は
それぞれ `(H,W)(軸 0=D を潰す)、(D,W)(軸 1=H を潰す)、(D,H)`(軸 2=W を潰す)、
dtype float64。値は入力の最大値そのまま(正規化しない)。負の値も max なので通り、
密度 0 の背景は 0 のまま。
形の検証は無い(2-D は `axis=2` で失敗し、4-D 以上は動いてしまう)ので、呼び手で次元を確かめる。
`match_mip_2d` はこの 3 枚に 2D NCC を掛けて 3D 位置を冗長推定する。任意視点の投影は
`render_volume_projection`(mode="mip")。
• 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.
• transforms_repr — py -3.11 examples_3d/transforms_repr.py
images as input)fuse_to_voxel · photometric_stereo · wrapped_phase · graycode_decode · decode_fringe · carve · visual_hull
transform)points_to_voxel · gaussians_to_voxel · mesh_to_voxel · mesh_to_points · depth_to_points · voxel_to_mesh · tsdf_from_depth · signed_distance_field
*Provenance: match3d.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.