typed op• 資料種類:video → video
• 呼叫:fullseye.apply(img, "tb_temporal_bilateral", a=0.5, b=0.5)(2-D 的模型是一張圖 + 兩個純量旋鈕 a,b∈[0,1])
*無圖: 該運算子以 video 為輸入。從影像開始的 Studio 程式無法到達該型別,請參見下方的可執行範例。*
逐幀因果時域雙邊降噪 → `(T, H, W)(video`)。
> 以下的詳細說明為原文 —— 摘要與標題已翻譯。
Edge-preserving in time: averages recent frames but drops the weight of
frames that differ (moved), so it denoises static regions without ghosting
the motion the way :func:moving_average_window does.
Typed bridge of the videostream op `temporal_bilateral into the 2-D evolution registry: the same implementation, called under the op(v, a, b) convention. a drives window (default 5) and b drives sigma_t` (default 2).
• 範例資料目錄(下載 URL / 授權) —— 2-D 用 skimage.data(BSD/公有領域)加合成圖,3-D 給出真實資料源(Stanford/PDS 等)的下載 URL。
• 運算子來歷與參考文獻 —— 該運算子族所依據的研究/方法出處。
• (尚無)
video 作為輸入)identity · tb_temporal_bandpass · tb_temporal_band_power · tb_temporal_median_window · tb_moving_average_window · tb_background_subtraction_window · tb_frame_difference_causal · tb_exponential_background
typed)tb_points_to_voxel · tb_estimate_point_normals · tb_iss_keypoints · tb_angle_3points · tb_project_points · tb_render_point_depth · tb_statistical_outlier_removal · tb_radius_outlier_removal
*Provenance: ops.py — 2D 運算子登記表。本條目由 tools/opdocs.py md 自動產生(請勿手動編輯)。*
© 2026 Kazufumi Furuse — Fullseye operator documentation. Licensed under Apache-2.0.