roadjp op• Data kinds: rgba → table
• Call: import fullseye as fs; fs.ledger.plate_mesh_from_image(rgba, width: 'float', height: 'float', *, cell: 'int' = 4, thickness: 'float' = 0.004, back_color=(0.55, 0.55, 0.55), label: 'int' = 4) -> 'dict' (to call the implementation directly, import roadjp; roadjp.plate_mesh_from_image(rgba, width: 'float', height: 'float', *, cell: 'int' = 4, thickness: 'float' = 0.004, back_color=(0.55, 0.55, 0.55), label: 'int' = 4) -> 'dict'; from the registry, opsdrive.get("plate_mesh_from_image"))
Turns an RGBA image into the mesh of a thin plate carrying the face colors.
The image is cut into tiles of cell × cell pixels, and only tiles whose mean alpha is 0.5 or more become faces (2 triangles). Face colour = the alpha-weighted mean colour of the tile Σ(α·rgb) / Σα. The front normal = −x (facing a car approaching in +x at yaw 0); the back has the same shape with back_color (normal +x). Origin = centre of the plate; the plate lies in the y-z plane (image left = +y, top = +z); the thickness thickness is spread ±half along x. Unused vertices are not kept (so that dimensions can be measured from the vertices).
Return value {"V", "F", "color", "label", "front_faces": (0, n), "back_faces": (n, 2n), "cells": (rows, cols)}. Empty input (rgba with 0 elements) raises ValueError.
• 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.
• gallery_drive_japan_roads — py -3.11 examples/gallery_drive_japan_roads.py
table as input)course_layout · course_occupancy · course_contains · world_build · world_camera · world_move · lidar_scan · rss_longitudinal_same
roadjp)sign_params · sign_image · sign_mesh · add_sign · signal_jp_mesh · add_signal_jp
*Provenance: roadjp.py — DRIVE 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.