fullseye

tb_quantize — 2D typed op

tb_quantize: input → output

図は合成の入力 128×128 で実際に走らせた出力。左が入力、右が出力。点群は上から見た散布(明るさ = z)、1-D 列は折れ線、体積は z 方向の最大値投影、動画は中央フレーム、複素画像は振幅、絵にならない返り値は値そのもの。

つまみ a を振る(0.1 / 0.5 / 0.9、もう一方は既定):

tb_quantize: knob a sweep

つまみ b は出力を変えない(実測: 0.1 / 0.5 / 0.9 で同一)。

段階(前置きの op → この op。左から順):

tb_quantize: stages

別の画像でも(合成シーン / 写真 / 硬貨。上段が入力、下段がその出力。つまみは既定):

tb_quantize: other inputs

使い方

Scalar quantiser with the error model stated, plus optional dither.

bits sets the number of levels L = 2**bits over the signal’s own min..max range. mode is "round" (mid-tread, unbiased) or "truncate" (floor, the convention PIL’s posterize and most fixed-point casts use).

The two differ by more than a rounding convention. With step Delta = range/(L-1) both have error variance Delta**2/12, but truncation also carries a mean of -Delta/2, so its mean square error is

truncate:  Delta**2/12 + (Delta/2)**2 = Delta**2/3
round:     Delta**2/12

— a factor of 4. Anything that measures a level (not just displays it) must round.

dither adds noise before quantising so the error stops being a function of the signal: "tpdf" (triangular, the audio standard — two uniform draws summed, so the error’s variance no longer depends on the sample value) or "rpdf" (one uniform draw). Dither raises the total error power but removes the correlation that makes quantisation audible as distortion rather than as hiss. seed fixes the draw so the op stays deterministic.

Applicability. (1) The range is taken from this signal, so two signals quantised separately do not share a scale. (2) bits=1 with no dither is a comparator, not a quantiser — the error model does not apply. (3) The error model assumes the signal moves by more than a step between samples; on a flat stretch the error is a constant offset, not noise.

Empty input (x with 0 elements) raises ValueError.

Typed bridge of the 1d op quantize into the 2-D evolution registry: the same implementation, called under the op(v, a, b) convention. a drives bits (default 8); b is unused.

参考(サンプルデータ・文献)

Studio で試す

下のプログラムは実際に走ることを確かめてある(図と同じ入力)。Studio のヘルプではこのブロックがボタンになり、その場で読み込んで実行できる。

img_to_signal 0.50 0.50
tb_quantize 0.50 0.50

実行できる例(この op を実際に呼ぶ検証済みサンプル)

次の例は元の台帳 op quantize を呼ぶもの。この橋渡し op は同じ実装を fn(v, a, b) 規約に合わせただけなので、挙動はそのまま当てはまる(呼び出し形だけ違う)。

型が繋がる次の op(signal を入力に取れる)

identity · tb_create_funct_1d_array · tb_smooth_funct_1d_gauss · tb_smooth_funct_1d_mean · tb_derivate_funct_1d · tb_integrate_funct_1d · tb_zero_crossings_funct_1d · tb_abs_funct_1d

同カテゴリ(typed)

tb_points_to_voxel · tb_estimate_point_normals · tb_iss_keypoints · tb_project_points · tb_render_point_depth · tb_points_render · tb_statistical_outlier_removal · tb_radius_outlier_removal


Provenance: ops.py — 2D operator registry. この per-op ノートは tools/opdocs.py md が自動生成(手編集しない)。

© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.