BriefGPT.xyz
Jul, 2016
未剪辑视频分类用于活动检测：参加ActivityNet挑战赛
Untrimmed Video Classification for Activity Detection: submission to ActivityNet Challenge
HTML
PDF
Gurkirt Singh, Fabio Cuzzolin
TL;DR
提出了一种新的方法，通过全局特征分析，预测未裁剪视频的前k个标签，并结合动态规划实现带时序约束的活动检测。该方法可用于未裁剪视频分类模型的时间检测。
Abstract
Current state-of-the-art
human activity recognition
is focused on the
classification
of temporally trimmed videos in which only one action occurs per frame. We propose a simple, yet effective, method for the
→