world optable × matrix × matrix → tableimport fullseye as fs; fs.ledger.world_camera(world: 'dict', pose, K, width: 'int' = 640, height: 'int' = 400, *, light=(0.3, -0.5, 0.8), ambient: 'float' = 0.35, sky=(0.62, 0.75, 0.92), ego=None) -> 'dict' (実装を直接呼ぶなら import driveworld; driveworld.world_camera(world: 'dict', pose, K, width: 'int' = 640, height: 'int' = 400, *, light=(0.3, -0.5, 0.8), ambient: 'float' = 0.35, sky=(0.62, 0.75, 0.92), ego=None) -> 'dict'、台帳から引くなら opsdrive.get("world_camera"))世界をカメラで撮る: {"color" (H,W,3), "label" (H,W) int(−1 = 空), "depth" (H,W), "face" (H,W), "shade" (H,W)}。
render3d.render_mesh(attributes=True) の三角形 id から面の色・ラベルを引き、色は Lambert
(ambient + (1−ambient)·max(n·l, 0)、n は render3d のカメラ系法線を世界系に戻さず、光をカメラ系で与える)。
ego=(V, F, color) を渡すと自車のメッシュも一緒に描く(世界には足さない)。
空の入力(world が空の dict)は ValueError。
py -3.11 examples/poc_ball_bounce.pypy -3.11 examples/poc_driving_crossing.pypy -3.11 examples/poc_driving_decisions.pypy -3.11 examples/poc_driving_endless_map.pypy -3.11 examples/poc_driving_humanoids.pypy -3.11 examples/poc_driving_japan_town.pypy -3.11 examples/poc_driving_lateral.pypy -3.11 examples/poc_driving_longitudinal.pypy -3.11 examples/poc_driving_pass.pypy -3.11 examples/poc_driving_school.pypy -3.11 examples/poc_driving_town.pypy -3.11 examples/poc_driving_traffic.pypy -3.11 examples/poc_kendama.pypy -3.11 examples/poc_table_tennis_bounce.pypy -3.11 examples/poc_table_tennis_spin.pypy -3.11 examples/poc_ttc_rss.pypy -3.11 examples/poc_world_terrain.pytable を入力に取れる)course_layout · course_occupancy · course_contains · world_build · world_move · lidar_scan · rss_longitudinal_same · rss_longitudinal_opposite
world)world_build · load_asset · world_move
Provenance: driveworld.py — DRIVE 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.