TGIF-QA

doi:doi:10.57702/exmq0j83

TGIF-QA

The TGIF-QA dataset consists of 165165 QA pairs chosen from 71741 animated GIFs. To evaluate the spatiotemporal reasoning ability at the video level, TGIF-QA dataset designs four unique task types, i.e., repetition count, repeating action, state transition and frame QA.

BibTex:

@dataset{Deng_Huang_and_Peihao_Chen_and_Runhao_Zeng_and_Qing_Du_and_Mingkui_Tan_and_Chuang_Gan_2024,
    abstract = {The TGIF-QA dataset consists of 165165 QA pairs chosen from 71741 animated GIFs. To evaluate the spatiotemporal reasoning ability at the video level, TGIF-QA dataset designs four unique task types, i.e., repetition count, repeating action, state transition and frame QA.},
    author = {Deng Huang and Peihao Chen and Runhao Zeng and Qing Du and Mingkui Tan and Chuang Gan},
    doi = {10.57702/exmq0j83},
    institution = {No Organization},
    keyword = {'Multimodal Learning', 'Question Answering', 'Spatio-temporal reasoning', 'VideoQA', 'Visual Question Answering', 'activity recognition', 'frame QA', 'natural language form', 'question answering', 'repeating action', 'repetition count', 'spatio-temporal reasoning', 'spatiotemporal reasoning', 'state transition', 'temporal modeling', 'video question answering', 'video understanding'},
    month = {dec},
    publisher = {TIB},
    title = {TGIF-QA},
    url = {https://service.tib.eu/ldmservice/dataset/tgif-qa},
    year = {2024}
}