Articulated pose estimation method research based on deep learning

  • Kai Chen
  • , Guitao Cao*
  • , Dan Meng
  • , Wenming Cao
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper proposes an articulated pose estimation method based on DeepPose, improving it's estimation accuracy. Firstly, it analyzes and compares the advantages and disadvantages of Faster Rcnn and DeepPose in the field of articulated pose estimation, proposes the R-Optimization algorithm to reajust joints' bounding box properly. Secondly, through combining DeepPose and Faster Rcnn, an improved articulated pose estimation method is proposed, which is better than using DeepPose or Faster Rcnn separately.

Original languageEnglish
Title of host publicationWMSCI 2017 - 21st World Multi-Conference on Systemics, Cybernetics and Informatics, Proceedings
EditorsNagib C. Callaos, Michael Savoie, Andres Tremante, Belkis Sanchez
PublisherInternational Institute of Informatics and Systemics, IIIS
Pages215-220
Number of pages6
ISBN (Electronic)9781941763599
StatePublished - 2017
Event21st World Multi-Conference on Systemics, Cybernetics and Informatics, WMSCI 2017 - Orlando, United States
Duration: 8 Jul 201711 Jul 2017

Publication series

NameWMSCI 2017 - 21st World Multi-Conference on Systemics, Cybernetics and Informatics, Proceedings
Volume1

Conference

Conference21st World Multi-Conference on Systemics, Cybernetics and Informatics, WMSCI 2017
Country/TerritoryUnited States
CityOrlando
Period8/07/1711/07/17

Keywords

  • Articulated pose estimation
  • Caffe
  • Convolution neural network
  • Deep learning

Fingerprint

Dive into the research topics of 'Articulated pose estimation method research based on deep learning'. Together they form a unique fingerprint.

Cite this