基于图形堆叠沙漏网络的三维人体姿态估计

Mar, 2021

基于图形堆叠沙漏网络的三维人体姿态估计

Graph Stacked Hourglass Networks for 3D Human Pose Estimation

Tianhan Xu, Wataru Takano

TL;DR本文提出了一种新的图卷积神经网络架构，用于2D到3D人体姿态估计任务，并使用多尺度、多级别特征表示的多尺度架构进行了实验，结果表明该模型优于现有技术，并取得了更好的表现。

Abstract

In this paper, we propose a novel graph convolutional network architecture, Graph Stacked Hourglass Networks, for 2D-to-3D human pose estimation tasks. The proposed architecture consists of repeated encoder-decod