BriefGPT.xyz
Sep, 2015
一种基于双源方法的单张图像3D姿势估计
3D Pose Estimation from a Single Monocular Image
HTML
PDF
Hashim Yasin, Umar Iqbal, Björn Krüger, Andreas Weber, Juergen Gall
TL;DR
通过两种独立的训练来源(2D姿势标注的图像和准确的3D运动捕捉数据)，提出一种双源方法，将2D姿态估计与高效稳健的3D姿态检索结合，从而解决单个RGB图像的3D姿态估计中获取足够训练数据的问题，实验表明，该方法达到了最先进的结果，并且即使是两个训练来源中的骨架结构差别显著时也具有竞争性。
Abstract
One major challenge for
3d pose estimation
from a single RGB image is the acquisition of sufficient
training data
. In particular, collecting large amounts of
→