BriefGPT.xyz
Nov, 2023
基于对象的(却不局限于类的)视频领域适应
Object-based (yet Class-agnostic) Video Domain Adaptation
HTML
PDF
Dantong Niu, Amir Bar, Roei Herzig, Trevor Darrell, Anna Rohrbach
TL;DR
视频领域适应的对象化(但类别不可知)视频领域适应(ODAPT)框架通过利用目标领域中一组稀疏的具有类别不可知对象注释的帧来适应现有的动作识别系统,为新领域的适应提供了一种简单而有效的方法。
Abstract
Existing
video-based action recognition
systems typically require dense annotation and struggle in environments when there is significant distribution shift relative to the training data. Current methods for
video domai
→