{
  "id": "2004.03044",
  "version": "v1",
  "published": "2020-04-06T23:58:59.000Z",
  "updated": "2020-04-06T23:58:59.000Z",
  "title": "When, Where, and What? A New Dataset for Anomaly Detection in Driving Videos",
  "authors": [
    "Yu Yao",
    "Xizi Wang",
    "Mingze Xu",
    "Zelin Pu",
    "Ella Atkins",
    "David Crandall"
  ],
  "comment": "23 pages, 11 figures, 6 tables",
  "categories": [
    "cs.CV"
  ],
  "abstract": "Video anomaly detection (VAD) has been extensively studied. However, research on egocentric traffic videos with dynamic scenes lacks large-scale benchmark datasets as well as effective evaluation metrics. This paper proposes traffic anomaly detection with a \\textit{when-where-what} pipeline to detect, localize, and recognize anomalous events from egocentric videos. We introduce a new dataset called Detection of Traffic Anomaly (DoTA) containing 4,677 videos with temporal, spatial, and categorical annotations. A new spatial-temporal area under curve (STAUC) evaluation metric is proposed and used with DoTA. State-of-the-art methods are benchmarked for two VAD-related tasks.Experimental results show STAUC is an effective VAD metric. To our knowledge, DoTA is the largest traffic anomaly dataset to-date and is the first supporting traffic anomaly studies across when-where-what perspectives. Our code and dataset can be found in: https://github.com/MoonBlvd/Detection-of-Traffic-Anomaly",
  "revisions": [
    {
      "version": "v1",
      "updated": "2020-04-06T23:58:59.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "anomaly detection",
      "driving videos",
      "supporting traffic anomaly studies",
      "dynamic scenes lacks large-scale benchmark",
      "scenes lacks large-scale benchmark datasets"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 23,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}