BriefGPT.xyz
Sep, 2021
WhyAct: 生活视频中的行动原因识别
WhyAct: Identifying Action Reasons in Lifestyle Vlogs
HTML
PDF
Oana Ignat, Santiago Castro, Hanwen Miao, Weiji Li, Rada Mihalcea
TL;DR
本文介绍一种利用视频和文本信息来自动识别在线视频中人类行为原因的多模态模型，并基于手动注释数据集WhyAct，研究了流行的生活方式视频类型。
Abstract
We aim to automatically identify
human action reasons
in
online videos
. We focus on the widespread genre of
lifestyle vlogs
, in which peop
→