BriefGPT.xyz
Mar, 2016
用于人体姿态估计的叠层沙漏网络
Stacked Hourglass Networks for Human Pose Estimation
HTML
PDF
Alejandro Newell, Kaiyu Yang, Jia Deng
TL;DR
该研究提出了一种新的卷积神经网络架构来进行人体姿势估计，采用自下而上、自上而下的处理方式及中间监督来提高网络的性能，在 FLIC 和 MPII 基准测试中取得了最先进的成果，并以堆叠沙漏网络总结了该架构的步骤。
Abstract
This work introduces a novel
convolutional network
architecture for the task of
human pose estimation
. Features are processed across all scales and consolidated to best capture the various spatial relationships a
→