Template based human pose and shape estimation from a single RGB-D image

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceedingpeer-review

Abstract

Estimating the 3D model of the human body is needed for many applications. However, this is a challenging problem since the human body inherently has a high complexity due to self-occlusions and articulation. We present a method to reconstruct the 3D human body model from a single RGB-D image. 2D joint points are firstly predicted by a CNN-based model called convolutional pose machine, and the 3D joint points are calculated using the depth image. Then, we propose to utilize both 2D and 3D joint points, which provide more information, to fit a parametric body model (SMPL). This is implemented through minimizing an objective function, which measures the difference of the joint points between the observed model and the parametric model. The pose and shape parameters of the body are obtained through optimization and the final 3D model is estimated. The experiments on synthetic data and real data demonstrate that our method can estimate the 3D human body model correctly.

Original languageEnglish
Title of host publicationICPRAM 2019 - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods
EditorsAna Fred, Maria De Marsico, Gabriella Sanniti di Baja
PublisherSciTePress
Pages574-581
Number of pages8
ISBN (Electronic)9789897583513
DOIs
Publication statusPublished - 2019
Event8th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2019 - Prague, Czech Republic
Duration: 2019 Feb 192019 Feb 21

Conference

Conference8th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2019
Country/TerritoryCzech Republic
CityPrague
Period2019/02/192019/02/21

Subject classification (UKÄ)

  • Computer Vision and Robotics (Autonomous Systems)

Free keywords

  • 2D and 3D Pose
  • Human Body Reconstruction
  • Pose
  • Shape Estimation
  • SMPL Model

Fingerprint

Dive into the research topics of 'Template based human pose and shape estimation from a single RGB-D image'. Together they form a unique fingerprint.

Cite this