arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1902.09774cs.CLcs.CV
收藏

可视化对话的图像-问题-回答协同网络

Image-Question-Answer Synergistic Network for Visual Dialog

UBTECH Sydney AI Centre(优必选悉尼人工智能中心);University of Sydney(悉尼大学)

Dalu Guo, Chang Xu, Dacheng Tao

摘要:

图像、问题(与历史相结合,用于解引用)和相应的答案是视觉对话的三个重要组成部分。经典的视觉对话系统将图像、问题和历史整合在一起来搜索或生成最佳匹配的答案,因此,这种方法明显忽略了答案的作用。在本文中,我们设计了一种新的图像-问题-答案协同网络,以评估答案在精确视觉对话中的作用。我们将传统的一阶段解决方案扩展为两阶段解决方案。在第一阶段中,候选答案根据其与图像和问题对的相关性进行粗略评分。然后,在第二阶段中,通过与图像和问题的协同作用,对正确概率高的答案进行重新排序。在Visual Dialog v1.0数据集上,所提出的协同网络增强了区分性视觉对话模型,以实现57.88%的归一化折扣累积增益的新的最先进水平。一个生成的可视化对话模型配备所提出的技术也显示出有前途的改进。

英文摘要:

The image, question (combined with the history for de-referencing), and the corresponding answer are three vital components of visual dialog. Classical visual dialog systems integrate the image, question, and history to search for or generate the best matched answer, and so, this approach significantly ignores the role of the answer. In this paper, we devise a novel image-question-answer synergistic network to value the role of the answer for precise visual dialog. We extend the traditional one-stage solution to a two-stage solution. In the first stage, candidate answers are coarsely scored according to their relevance to the image and question pair. Afterward, in the second stage, answers with high probability of being correct are re-ranked by synergizing with image and question. On the Visual Dialog v1.0 dataset, the proposed synergistic network boosts the discriminative visual dialog model to achieve a new state-of-the-art of 57.88\% normalized discounted cumulative gain. A generative visual dialog model equipped with the proposed technique also shows promising improvements.

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。

↑