UBnormal

doi:doi:10.57702/iyiaephv

UBnormal

The UBnormal dataset is a recent and large dataset proposed for supervised anomaly detection. In contrast to other benchmarks, it is a synthetic dataset created with Cinema4D. Starting from 29 natural 2D images with different backgrounds, the authors obtain virtual 3D scenes. For each scene, they produced 19 clips, which encompass both normal and abnormal events.

BibTex:

@dataset{Andra_Acsintoae_and_Andrei_Florescu_and_Mariana-Iuliana_Georgescu_and_Tudor_Mare_and_Paul_Sumedrea_and_Radu_Tudor_Ionescu_and_Fahad_Shahbaz_Khan_and_Mubarak_Shah_2024,
    abstract = {The UBnormal dataset is a recent and large dataset proposed for supervised anomaly detection. In contrast to other benchmarks, it is a synthetic dataset created with Cinema4D. Starting from 29 natural 2D images with different backgrounds, the authors obtain virtual 3D scenes. For each scene, they produced 19 clips, which encompass both normal and abnormal events.},
    author = {Andra Acsintoae and Andrei Florescu and Mariana-Iuliana Georgescu and Tudor Mare and Paul Sumedrea and Radu Tudor Ionescu and Fahad Shahbaz Khan and Mubarak Shah},
    doi = {10.57702/iyiaephv},
    institution = {No Organization},
    keyword = {'Background Clutter', 'Cinema4D', 'Object Detection', 'Video Anomaly Detection', 'supervised anomaly detection', 'synthetic dataset'},
    month = {dec},
    publisher = {TIB},
    title = {UBnormal},
    url = {https://service.tib.eu/ldmservice/dataset/ubnormal},
    year = {2024}
}