Answer Distillation for Visual Question Answering
Fang, Zhiwei1,2; Liu, Jing1; Tang, Qu1; Li, Yong3; Lu, Hanqing1
2019-05
会议名称Asian Conference on Computer Vision
会议日期2018.12
会议地点Perth, Australia
摘要

Answering open-ended questions in Visual Question Answering (VQA) is a challenging task. As the answers are totally free-form, the answer space for open-ended questions is in nite in theory. This increases the diffculty for algorithms to predict the correct answers. In this paper, we propose a method named answer distillation to decrease the scale of answer space and limit the correct result into a small set of answer candidates. Speci cally, we design a two-stage architecture to answer a question: First, we develop an answer distillation network to distill the answers, converting an open-ended question to a multiple-choice one with a short list of answer candidates. Then, we make full use of the knowledge from the answer candidates to guide the visual attention and re ne the prediction results. Extensive experiments are conducted to validate the effiectiveness of our answer distillation architecture. The results show that our method can effiectively compress the answer space and improve the accuracy on open-ended task, providing a new state-of-the-art performance on COCO-VQA dataset.

语种英语
七大方向——子方向分类多模态智能
文献类型会议论文
条目标识符http://ir.ia.ac.cn/handle/173211/23599
专题模式识别国家重点实验室_图像与视频分析
通讯作者Liu, Jing
作者单位1.Institute of Automation, Chinese Academy of Sciences, Beijing, China
2.University of Chinese Academy of Sciences, Beijing, China
3.Business Growth BU, JD.com
第一作者单位中国科学院自动化研究所
通讯作者单位中国科学院自动化研究所
推荐引用方式
GB/T 7714
Fang, Zhiwei,Liu, Jing,Tang, Qu,et al. Answer Distillation for Visual Question Answering[C],2019.
条目包含的文件 下载所有文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
Answer Distillation (2077KB)会议论文 开放获取CC BY-NC-SA浏览 下载
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[Fang, Zhiwei]的文章
[Liu, Jing]的文章
[Tang, Qu]的文章
百度学术
百度学术中相似的文章
[Fang, Zhiwei]的文章
[Liu, Jing]的文章
[Tang, Qu]的文章
必应学术
必应学术中相似的文章
[Fang, Zhiwei]的文章
[Liu, Jing]的文章
[Tang, Qu]的文章
相关权益政策
暂无数据
收藏/分享
文件名: Answer Distillation for Visual Question Answering.pdf
格式: Adobe PDF
此文件暂不支持浏览
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。