Multiple Temporal Fusion based Weakly-supervised Pre-training Techniques for Video Categorization

Xiaochen Cai, Hengxing Cai, Boqing Zhu, Kele Xu, Weiwei Tu, Dawei Feng

Abstract

In this paper, we present our solution of the ACM Multimedia 2022 pre-training for video understanding challenge. First, we pre-train the models on large-scale weakly-supervised video datasets with different temporal resolutions, then fine-tune the model for downstream application. Quantitative comparisons are conducted to evaluate the performance of different networks at multiple temporal resolutions. Moreover, we fusion different pre-trained models through weighted averaging. We achieve an accuracy of 62.39% in the testing set, which ranked as the first place in the video categorization track of this challenge.
BibTeX
@inproceedings{cai2022multiple,
  author = {Xiaochen Cai and Hengxing Cai and Boqing Zhu and Kele Xu and Weiwei Tu and Dawei Feng},
  title = {{Multiple Temporal Fusion based Weakly-supervised Pre-training Techniques for Video Categorization}},
  booktitle = {ACM Multimedia},
  year = {2022},
  doi = {10.1145/3503161.3551585},
}

Loading related content...

Figure