ChinaOpen: A Dataset for Open-world Multimodal Learning

被引:0
|
作者
Chen, Aozhu [1 ]
Wang, Ziyuan [1 ]
Dong, Chengbo [1 ]
Tian, Kaibin [1 ]
Zhao, Ruixiang [1 ]
Liang, Xun [1 ]
Kang, Zhanhui [2 ]
Li, Xirong [3 ]
机构
[1] Renmin Univ China, Sch Informat, Beijing, Peoples R China
[2] Tencent, Shenzhen, Peoples R China
[3] Renmin Univ China, MoE Key Lab DEKE, Beijing, Peoples R China
关键词
Chinese video dataset; multimodal learning; multi-task evaluation;
D O I
10.1145/3581783.3612156
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper introduces ChinaOpen, a dataset sourced from Bilibili, a popular Chinese video-sharing website, for open-world multimodal learning. While the state-of-the-art multimodal learning networks have shown impressive performance in automated video annotation and cross-modal video retrieval, their training and evaluation are primarily conducted on YouTube videos with English text. Their effectiveness on Chinese data remains to be verified. In order to support multimodal learning in the new context, we construct ChinaOpen-50k, a webly annotated training set of 50k Bilibili videos associated with user-generated titles and tags. Both text-based and content-based data cleaning are performed to remove low-quality videos in advance. For a multi-faceted evaluation, we build ChinaOpen-1k, a manually labeled test set of 1k videos. Each test video is accompanied with a manually checked user title and a manually written caption. Besides, each video is manually tagged to describe objects / actions / scenes shown in the visual content. The original user tags are also manually checked. Moreover, with all the Chinese text translated into English, ChinaOpen-1k is also suited for evaluating models trained on English data. In addition to ChinaOpen, we propose Generative Video-to-text Transformer (GVT) for Chinese video captioning. We conduct an extensive evaluation of the state-of-the-art single-task / multi-task models on the new dataset, resulting in a number of novel findings and insights.
引用
收藏
页码:6432 / 6440
页数:9
相关论文
共 50 条
  • [1] OpenWGL: Open-World Graph Learning
    Wu, Man
    Pan, Shirui
    Zhu, Xingquan
    [J]. 20TH IEEE INTERNATIONAL CONFERENCE ON DATA MINING (ICDM 2020), 2020, : 681 - 690
  • [2] Open-World Lifelong Graph Learning
    Hoffmann, Marcel
    Galke, Lukas
    Scherp, Ansgar
    [J]. 2023 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, IJCNN, 2023,
  • [3] Challenges, evaluation and opportunities for open-world learning
    Kejriwal, Mayank
    Kildebeck, Eric
    Steininger, Robert
    Shrivastava, Abhinav
    [J]. NATURE MACHINE INTELLIGENCE, 2024, 6 (06) : 580 - 588
  • [4] Open-World Learning for Traffic Scenarios Categorisation
    Balasubramanian, Lakshman
    Wurst, Jonas
    Botsch, Michael
    Deng, Ke
    [J]. IEEE TRANSACTIONS ON INTELLIGENT VEHICLES, 2023, 8 (05): : 3506 - 3521
  • [5] Challenges, evaluation and opportunities for open-world learning
    Kejriwal, Mayank
    Kildebeck, Eric
    Steininger, Robert
    Shrivastava, Abhinav
    [J]. NATURE MACHINE INTELLIGENCE, 2024, : 580 - 588
  • [6] Open-world Learning and Application to Product Classification
    Xu, Hu
    Liu, Bing
    Shu, Lei
    Yu, P.
    [J]. WEB CONFERENCE 2019: PROCEEDINGS OF THE WORLD WIDE WEB CONFERENCE (WWW 2019), 2019, : 3413 - 3419
  • [7] Open-World Learning for Radically Autonomous Agents
    Langley, Pat
    [J]. THIRTY-FOURTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, THE THIRTY-SECOND INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE AND THE TENTH AAAI SYMPOSIUM ON EDUCATIONAL ADVANCES IN ARTIFICIAL INTELLIGENCE, 2020, 34 : 13539 - 13543
  • [8] Learning to Prompt Knowledge Transfer for Open-World Continual Learning
    Li, Yujie
    Yang, Xin
    Wang, Hao
    Wang, Xiangkun
    Li, Tianrui
    [J]. THIRTY-EIGHTH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, VOL 38 NO 12, 2024, : 13700 - 13708
  • [9] Open-world Machine Learning: Applications, Challenges, and Opportunities
    Parmar, Jitendra
    Chouhan, Satyendra
    Raychoudhury, Vaskar
    Rathore, Santosh
    [J]. ACM COMPUTING SURVEYS, 2023, 55 (10)
  • [10] Learning Refined Features for Open-World Text Classification
    Li, Zeting
    Cai, Yi
    Tan, Xingwei
    Han, Guoqiang
    Ren, Haopeng
    Wu, Xin
    Li, Wen
    [J]. WEB AND BIG DATA, APWEB-WAIM 2021, PT I, 2021, 12858 : 367 - 381