Torso orientation: A new clue for occlusion-aware human pose estimation

Yang Yu, Baoyao Yang, Pong Chi YUEN*

*Corresponding author for this work

Research output: Chapter in book/report/conference proceedingConference contributionpeer-review

Abstract

Self-occlusion is a challenging problem existing in human pose estimation. In this paper we exploit a new cue to solve this problem: the torso orientation. We describe a technique to automatically detect self-occlusion in training set without visibility label. Given this prior information, we are able to jointly learn an occlusion-aware model to capture the pattern of self-occluded body parts. We evaluate our model on two major datasets, which are both publicly available. The experiment result shows that our model is quite competitive in both of the datasets with the state-of-the-arts. By this way, we illustrate our model's robustness to the self-occlusion problem in human pose estimation.

Original languageEnglish
Title of host publication2016 24th European Signal Processing Conference, EUSIPCO 2016
PublisherEuropean Signal Processing Conference, EUSIPCO
Pages908-912
Number of pages5
ISBN (Electronic)9780992862657
DOIs
Publication statusPublished - 28 Nov 2016
Event24th European Signal Processing Conference, EUSIPCO 2016 - Budapest, Hungary
Duration: 28 Aug 20162 Sep 2016

Publication series

NameEuropean Signal Processing Conference
Volume2016-November
ISSN (Print)2219-5491

Conference

Conference24th European Signal Processing Conference, EUSIPCO 2016
Country/TerritoryHungary
CityBudapest
Period28/08/162/09/16

Scopus Subject Areas

  • Signal Processing
  • Electrical and Electronic Engineering

User-Defined Keywords

  • Articulated model
  • Computer vision
  • Pose estimation
  • Self-occlusion

Fingerprint

Dive into the research topics of 'Torso orientation: A new clue for occlusion-aware human pose estimation'. Together they form a unique fingerprint.

Cite this