HMDB-51

doi:doi:10.57702/6fpocbn8

HMDB-51

Motion has shown to be useful for video understanding, where motion is typically represented by optical flow. However, computing flow from video frames is very time-consuming. Recent works directly leverage the motion vectors and residuals readily available in the compressed video to represent motion at no cost.

BibTex:

@dataset{Hildegard_Kuehne_and_Hueihan_Jhuang_and_Est´ıbaliz_Garrote_and_Tomaso_Poggio_and_Thomas_Serre_2024,
    abstract = {Motion has shown to be useful for video understanding, where motion is typically represented by optical flow. However, computing flow from video frames is very time-consuming. Recent works directly leverage the motion vectors and residuals readily available in the compressed video to represent motion at no cost.},
    author = {Hildegard Kuehne and Hueihan Jhuang and Est´ıbaliz Garrote and Tomaso Poggio and Thomas Serre},
    doi = {10.57702/6fpocbn8},
    institution = {No Organization},
    keyword = {'HMDB', 'HMDB-51', 'Human Action Recognition', 'Video Understanding', 'action categories', 'action recognition', 'dataset', 'human action', 'human action recognition', 'human actions', 'human motion', 'human motion recognition', 'realistic videos', 'recognition', 'source spanning', 'video action classification', 'video action recognition', 'video recognition', 'video understanding', 'zero-shot learning'},
    month = {dec},
    publisher = {TIB},
    title = {HMDB-51},
    url = {https://service.tib.eu/ldmservice/dataset/hmdb-51},
    year = {2024}
}