基于深度学习的2D人体姿态估计浅析
Analysis of 2D Human Pose Estimation Based on Deep Learning
任肖月 1王俊英 1陈晗晗1
作者信息
- 1. 三峡大学水电工程智能视觉监测湖北省重点实验室;三峡大学湖北省建筑质量检测装备工程技术研究中心;三峡大学计算机与信息学院,湖北宜昌 443002
- 折叠
摘要
近年来,随着深度学习技术的发展,基于深度学习的2D人体姿态估计展现出了极其优异的效果.文章主要对2014年以来基于深度学习的2D人体姿态估计相关内容进行了分析和评述,包括对主流方法的介绍和对数据集评价指标的总结对比,最后还讨论了 2D人体姿态估计的未来研究方向.
Abstract
In recent years,with the development of deep learning technology,2D human pose es-timation based on deep learning has shown excellent results.This paper mainly describes the rel-evant content of 2D human pose estimation based on deep learning since 2014,including the in-troduction of mainstream methods and the summary and comparison of dataset evaluation indi-cators,and finally looks forward to the future of human pose estimation based on our own un-derstanding.
关键词
人体姿态估计/深度学习/关键点检测/神经网络Key words
human position estimation/deep learning/joints detection/neural networks引用本文复制引用
出版年
2024
NETL
NSTL
万方数据