Traffic Anomaly Reasoning (TAR)

Traffic Anomaly Reasoning (TAR) This dataset is the official release for AI City Challenge 2026 Track 3 — Anomalous Events in Transportation. It contains: Training (train/): 44,040 pseudo-labeled multi-task annotations covering 3,670 CCTV videos (≈26.1 hours: 9.2 hrs anomalous + 16.9 hrs normal) sourced from eight public datasets. The corresponding training videos are included in this dataset package under videos/. Test (test/): 960 human-curated annotations covering 80 short… See the full description on the dataset page: Track 3.

Source: Hugging Face Hub (Tohrumi/AICityChallenge_Track_3). Metadata imported from the dataset’s Hub tags.