arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2104.09411cs.CVcs.MM
收藏

基于对比多模态预训练的中文视频和语言理解

Understanding Chinese Video and Language via Contrastive Multimodal Pre-Training

University of Science and Technology of China(中国科学技术大学);Alibaba Group(阿里巴巴集团);Alibaba-NTU Singapore Joint Research Institute, Nanyang Technological University(阿里巴巴-南洋理工大学新加坡联合研究院);School of Computer Science and Engineering, Nanyang Technological University(南洋理工大学计算机科学与工程学院);Zhejiang University(浙江大学)

Chenyi Lei, Shixian Luo, Yong Liu, Wanggui He, Jiamang Wang, Guoxin Wang, Haihong Tang, Chunyan Miao, Houqiang Li

摘要:

预训练的神经模型最近在理解多模态内容方面取得了令人印象深刻的表现。然而,由于以下原因,预训练用于视频和语言理解的神经模型仍然非常具有挑战性,特别是对于中文视频语言数据。首先,现有的视频语言预训练算法主要关注单词和视频帧的共现,而忽略了视频语言内容的其他有价值的语义和结构信息,例如,时序和时空关系。其次,视频句子对齐与其他代理任务之间存在冲突。第三,缺乏大规模和高质量的中文视频语言数据集(例如,包括1000万个独特的视频),这是预训练技术成功的基本条件。在这项工作中,我们提出了一个新的视频-语言理解框架命名为VICTOR,这代表视频-语言理解通过对比multiTimOdal预训练。VICTOR在对比学习范式下构建了多个新的代理任务,使模型具有更强的鲁棒性,能够从不同角度捕捉更复杂的多模态语义和结构关系。VICTOR是在一个大规模的中文视频语言数据集上训练的,包括超过1000万个完整的视频,以及相应的高质量文本描述。我们将预训练的VICTOR模型应用于一系列下游应用,并与最先进的预训练方法(如VideoBERT和UniVL)进行比较,展示了其优越的性能。这些守则和经过培训的检查站将向公众开放,以促进研究界的进一步发展。

英文摘要:

The pre-trained neural models have recently achieved impressive performances in understanding multimodal content. However, it is still very challenging to pre-train neural models for video and language understanding, especially for Chinese video-language data, due to the following reasons. Firstly, existing video-language pre-training algorithms mainly focus on the co-occurrence of words and video frames, but ignore other valuable semantic and structure information of video-language content, e.g., sequential order and spatiotemporal relationships. Secondly, there exist conflicts between video sentence alignment and other proxy tasks. Thirdly, there is a lack of large-scale and high-quality Chinese video-language datasets (e.g., including 10 million unique videos), which are the fundamental success conditions for pre-training techniques. In this work, we propose a novel video-language understanding framework named VICTOR, which stands for VIdeo-language understanding via Contrastive mulTimOdal pRe-training. Besides general proxy tasks such as masked language modeling, VICTOR constructs several novel proxy tasks under the contrastive learning paradigm, making the model be more robust and able to capture more complex multimodal semantic and structural relationships from different perspectives. VICTOR is trained on a large-scale Chinese video-language dataset, including over 10 million complete videos with corresponding high-quality textual descriptions. We apply the pre-trained VICTOR model to a series of downstream applications and demonstrate its superior performances, comparing against the state-of-the-art pre-training methods such as VideoBERT and UniVL. The codes and trained checkpoints will be publicly available to nourish further developments of the research community.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑