BriefGPT.xyz
Jun, 2015
视频人体姿势估计的流动卷积神经网络
Flowing ConvNets for Human Pose Estimation in Videos
HTML
PDF
Tomas Pfister, James Charles, Andrew Zisserman
TL;DR
本文提出了一种基于深度学习和光流的人体姿势估计算法，通过引入时间序列信息和光流辅助，该算法在多个姿态估计数据集上实现了超越当前最先进方法的性能。
Abstract
The objective of this work is
human pose estimation
in videos, where multiple frames are available. We investigate a
convnet architecture
that is able to benefit from
→