BriefGPT.xyz
Mar, 2018
RGBD图像中的人体姿态三维估计用于机器人任务学习
3D Human Pose Estimation in RGBD Images for Robotic Task Learning
HTML
PDF
Christian Zimmermann, Tim Welschehold, Christian Dornhege, Wolfram Burgard, Thomas Brox
TL;DR
利用RGBD图像估算3D人体姿态，通过关键点检测器和深度信息实现3D提升，在真实环境下通过学习演示框架指示服务机器人模仿人类教师的操作，超越了单眼调色板和深度姿态估计方法的性能。
Abstract
We propose an approach to estimate
3d human pose
in real world units from a single
rgbd image
and show that it exceeds performance of monocular 3D pose estimation approaches from color as well as pose estimation
→