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 1907.03049cs.CVcs.CL
收藏

基于跨模态自注意网络学习的视频问题生成

Video Question Generation via Cross-Modal Self-Attention Networks Learning

University of Toronto(多伦多大学);National Taiwan University(台湾大学)

Yu-Siang Wang, Hung-Ting Su, Chen-Hsi Chang, Zhe-Yu Liu, Winston H. Hsu

摘要:

我们介绍了一个新的任务,视频问题生成(视频QG)。视频QG模型自动生成问题,给出视频剪辑及其相应的对话。视频QG需要一系列技能-句子理解,时间关系,视觉和语言之间的相互作用,以及提出有意义问题的能力。为了解决这个问题,我们提出了一种新的语义丰富的跨模态自我注意(SRCMSA)网络聚合的多模态和不同的功能。更准确地说,我们通过整合对象级信息来增强视频帧的语义,并且我们共同考虑了视频问题生成任务的跨模态注意力。令人兴奋的是,我们提出的模型显着提高了TVQA数据集上BLEU-4得分的基线从7.58到14.48。最重要的是,我们可以说是为理解具有挑战性的视频输入铺平了一条新的道路,我们提供了详细的多样性分析,为未来的研究开辟了道路。

英文摘要:

We introduce a novel task, Video Question Generation (Video QG). A Video QG model automatically generates questions given a video clip and its corresponding dialogues. Video QG requires a range of skills -- sentence comprehension, temporal relation, the interplay between vision and language, and the ability to ask meaningful questions. To address this, we propose a novel semantic rich cross-modal self-attention (SRCMSA) network to aggregate the multi-modal and diverse features. To be more precise, we enhance the video frames semantic by integrating the object-level information, and we jointly consider the cross-modal attention for the video question generation task. Excitingly, our proposed model remarkably improves the baseline from 7.58 to 14.48 in the BLEU-4 score on the TVQA dataset. Most of all, we arguably pave a novel path toward understanding the challenging video input and we provide detailed analysis in terms of diversity, which ushers the avenues for future investigations.

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。

↑