volseq_magnify_motion — LIVE4D time op

• Data kinds: volseq → volseq

• Call: import fullseye as fs; fs.ledger.volseq_magnify_motion(volseq, alpha: 'float' = 10.0, f_lo: 'float' = 0.05, f_hi: 'float' = 0.4, fps: 'float' = 1.0, sigma: 'float' = 0.5) -> 'np.ndarray' (to call the implementation directly, import live4d; live4d.volseq_magnify_motion(volseq, alpha: 'float' = 10.0, f_lo: 'float' = 0.05, f_hi: 'float' = 0.4, fps: 'float' = 1.0, sigma: 'float' = 0.5) -> 'np.ndarray'; from the registry, opslive4d.get("volseq_magnify_motion"))

Usage

> This operator's description has not been translated yet. The original text follows as it is.

体積時系列の帯域内の微小な動きを alpha 倍にする(`volseq`、Eulerian の線形拡大の 3 次元版)。

Wu et al.(SIGGRAPH 2012)の線形 Eulerian: 各体積を空間ガウス(`sigma`)で平滑化した層を、時間方向に

帯域通過(`[f_lo, f_hi] Hz、fps は体積のフレームレート、FFT の理想帯域)して alpha − 1` 倍を

元に足し戻す。並進 δ の像 `I(x − δ(t)) を 1 次で展開すると帯域内の時間変化は −δ · ∇I` なので、

足し戻した結果の変位は `alpha · δ(alpha` は変位の倍率: 1 で恒等、0 で帯域内の動きを消す)。

成り立つのは 小さな動きだけ —— 目安は `alpha · δ < λ / 8`(λ は平滑化後の空間波長 ≈ 4·sigma)。

それを超えると像が壊れる(増幅でなく歪み)。雑音も同じ倍率で増幅される(SNR は良くならない)。

>>> s = volseq_synth_beating((16, 24, 24), n_frames=24, period=8.0, amplitude=0.1)

>>> volseq_magnify_motion(s, alpha=8.0, f_lo=0.08, f_hi=0.2, fps=1.0).shape

(24, 16, 24, 24)

Detailed usage guide

• live4d family guide

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.

• The canonical algorithm (author, year) and its uses are named in the family usage guide above.

Runnable examples (verified samples that actually call this op)

• poc_live4d — py -3.11 examples/poc_live4d.py

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

volseq_mip_video · volseq_cut_video · volseq_speed · volseq_pathline_render · volseq_pathline_orbit · volseq_interpolate_flow · volseq_render_orbit · focus_sweep_height_video

Same category (time)

video_interpolate_flow · volseq_interpolate_flow

*Provenance: live4d.py — LIVE4D 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.