BriefGPT.xyz
Apr, 2021
Skeletor: 基于骨架的Transformer模型用于鲁棒的身体姿态估计
Skeletor: Skeletal Transformers for Robust Body-Pose Estimation
HTML
PDF
Tao Jiang, Necati Cihan Camgoz, Richard Bowden
TL;DR
本篇研究旨在解决从一张单目视频中预测3D人体姿态的问题,提出了一种基于Transformer的网络Skeletor,通过无监督学习学习运动的三维姿态与运动分布,有效克服低分辨率、模糊运动和部分骨骼损坏等问题,同时提高了3D人体姿态估计的性能。
Abstract
Predicting
3d human pose
from a single
monoscopic video
can be highly challenging due to factors such as low resolution, motion blur and occlusion, in addition to the fundamental ambiguity in estimating 3D from 2
→