emb-ai/traffic-sign-bench

Traffic Sign Bench

Official per-sign SUMO maps for TrafficRuleBench: real Moscow OSM layouts, 25 signs, 2500 maps. Protocol size is 80 train + 20 test maps per sign.

Road geometry is derived from OpenStreetMap © OpenStreetMap contributors and is released under ODbL 1.0.

Download

All scenes land under data/scenes/<sign>/<scene_id>/, which is what eval expects:

huggingface-cli download emb-ai/traffic-sign-bench \
    --repo-type dataset \
    --local-dir data

Code and runners: https://github.com/emb-ai/traffic-rule-bench.

from datasets import load_dataset
ds = load_dataset("emb-ai/traffic-sign-bench", split="train")

That loads the catalog (one row per scene, with a path to the preview and the net). The .net.xml files themselves are next to the catalog in scenes/.

Layout

scenes/<sign>/<scene_id>/
  map.net.xml
  meta.json
  custom_cropped.png
metadata/
  signs.yaml
  sign_allocations.json
  train_ids.json
  test_ids.json
  catalog.jsonl
  catalog.parquet

sign is the eval id (yield, not 2.4). Train/test is in sign_allocations.json and the catalog, not in the folder name.

Signs

Sign	PDD	Family	Train	Test	Total
blocked_road	3.2	junction	80	20	100
crosswalk	5.19	segment	80	20	100
detour_either	4.2.3	segment	80	20	100
detour_left	4.2.2	segment	80	20	100
detour_right	4.2.1	segment	80	20	100
direction_left	4.1.3	dual_path	80	20	100
direction_left_right	4.1.6	dual_path	80	20	100
direction_right	4.1.2	dual_path	80	20	100
direction_straight	4.1.1	dual_path	80	20	100
direction_straight_left	4.1.5	dual_path	80	20	100
direction_straight_right	4.1.4	dual_path	80	20	100
main	2.1	junction	80	20	100
min_speed	4.6	segment	80	20	100
no_entry	3.1	dual_path	80	20	100
no_turn_left	3.18.2	dual_path	80	20	100
no_turn_right	3.18.1	dual_path	80	20	100
one_way_left	5.7.2	dual_path	80	20	100
one_way_right	5.7.1	dual_path	80	20	100
residential_zone	5.21	segment	80	20	100
roundabout	4.3	junction	80	20	100
secondary	2.3	junction	80	20	100
speed_limit	3.24	segment	80	20	100
stop	2.5	junction	80	20	100
yield	2.4	junction	80	20	100
zone_speed_limit	5.31	segment	80	20	100

How it was built

Moscow OSM → city SUMO net → junction / dual-path / segment crops → each sign queries that shared pool (signs.yaml) → materialize into data/scenes/<sign>/. Pedestrian crossings (PDD 5.19) get a mid-block zebra injected after copy.

Split is stamped on place identity before allocation:

• junctions and dual-path: junction_id
• segments: osm_way_id (one street does not appear in both halves)

Signs share the pool: the same map may be assigned to more than one sign.

Limitations

• Geography is Moscow only.
• On some 5.19 segments SUMO omits the crossing edge; the split node and sidewalks are still there.
• These folders are maps, not closed-loop eval manifests.

Citation

@misc{traffic-sign-bench,
  title  = {Traffic Sign Bench},
  author = {EMB AI},
  year   = {2026},
  url    = {https://huggingface.co/datasets/emb-ai/traffic-sign-bench}
}