sobel3d — 3D feature op

Data kinds: voxelgradient

Call: import fullseye as fs; fs.ledger.sobel3d(vol, device='cpu') (to call the implementation directly, import match3d; match3d.sobel3d(vol, device='cpu'); from the registry, ops3d.get("sobel3d"))

Return value through the ledger: fullseye.ledger.sobel3d(...) returns **only the declared out type gradient** (the underlying function also returns auxiliary values). When you need what was dropped, use fullseye.ledger.sobel3d.raw(...) or call match3d.sobel3d directly.

GPU: this op has a GPU path (device="cuda")

Usage

The 3-D gradient (gz, gy, gx). Separable conv3d of the derivative [-1,0,1] × the smoother [1,2,1].

> The detailed description below is the original text — the summary and the headings are translated.

vol は numpy でも torch tensor(GPU 上でも可)でも受ける(scene_flow 等の device 常駐用)。

返り値は torch tensor 3 本(numpy ではない)、各 `(1,1,D,H,W) float32、device` 上。

`gz は軸 0 方向、gy は軸 1、gx` は軸 2 の微分(入力が (D,H,W) なら (depth,row,col)

順)。numpy に戻すなら `g[0, 0].cpu().numpy()`。

利得: 微分 [-1,0,1] (傾き 1 で 2)× 他 2 軸の平滑 [1,2,1] (各 4)で 真の勾配の 32 倍が出る

(正規化しない)。真の値が要るなら 32 で割る(`curvature_maps / scene_flow_lk` は内部で

割っている。`hessian3d` は利得 1 なので混ぜるときに注意)。

端は replicate padding(境界で偽のエッジを作らない)。tensor 入力が 3-D なら batch 次元を

足し、5-D ならそのまま使う。dtype は float32 に落とす。

後段: `curvature_maps / match_shape_3d の単位勾配、hough_plane_3d` の法線。

方向の要らないエッジ強度なら `morph_gradient3d` も代替。

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)

diff_featurespy -3.11 examples_3d/diff_features.py

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

fuse_to_voxel

Same category (feature)

hessian3d · curvature_maps · edt_jfa · vol_frangi · vol_local_std · vol_local_thickness · vol_orientation_coherence · vol_euler_number


*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.