Special Issue Description and Topics

With the rapid development of computing and sensing technologies, human pose recovery and behaviour analysis are becoming emergent, and a lot of approaches have been proposed for potential applications, such as home entertainment, media content analysis, public surveillance/security, and education purposes. Machine learning and signal processing are critically important for precisely recovering human poses and effectively analysing behaviours.

This special issue aims to demonstrate how machine learning algorithms and signal processing techniques have contributed, and are contributing to the research and applications of human pose recovery and behaviour analysis. Many machine learning algorithms and signal processing techniques have already been successfully applied in this research area, e.g., supervised distance metric learning has been successfully used to estimate the dissimilarity of 3D poses; linear and multilinear dimension reduction methods efficiently enhance the human behavior recognition rates; sparse coding algorithm has shown its excellent performance in recovering 3D human poses from monocular images; graph cut has been applied to the problem of human body segmentation with high success; kernel machines, such as the Relevance Vector Machine and Support Vector Machine, are successfully used in recovering poses from image silhouettes; manifold learning algorithms are applied in motion synthesis.

The editors hope to collect a group of research results, which can report the recent developments in the related research topics. In addition, the researchers can exchange their innovative ideas on machine learning algorithms and signal processing techniques in human pose recovery and behaviour analysis by submitting manuscripts to this special issue. To summarize, this special issue prefers a large scope of submissions, which develop and adopt machine learning algorithms and signal processing techniques for human pose recovery and behaviour analysis. We are especially interested in 1) theoretical advances as well as algorithm developments in machine learning algorithms and signal processing techniques for specific human pose recovery and behaviour analysis problems, 2) useful applications and system creations in human pose recovery and behaviour analysis, and 3) new datasets as benchmarks for new progress. The following list contains topics of interest (but not limited to):

Distance metric learning for pose dissimilarity estimation
Intelligent home entertainment systems
Intelligent visual surveillance
Manifold learning for motion synthesis
Multimodal learning based human body descriptors
Probabilistic graphical models for image analysis and modeling
Semi-supervised learning based human body detection
Sparse learning and structured sparsity for pose recovery
Subspace learning methods for human behavior recognition
Unsupervised human skeleton extraction
Important dates:

Manuscript submission: March 01, 2014
Revision submission: July 01, 2014
Acceptance notification: August 15, 2014
Final manuscripts due: October 01, 2014
Anticipated publication: January 01 2015

Submission:

Manuscripts (Please follow Signal Processing publishing format, details can be found at http://www.elsevier.com/journals/signal-processing/0165-1684/guide-for-authors) should be submitted via the Electronic Editorial System, Elsevier: http://ees.elsevier.com/sigpro/ and please select “SI:MLSP for HPRBA” when you reach the “Article Type” step in the submission process.

Guide for authors can be found at: http://support.elsevier.com/app/answers/detail/a_id/116

Guest Editors:

Jun Yu
Xiamen University, Xiamen, China
zju.yujun@gmail.com

Huiyu Zhou
Queen's University Belfast, United Kingdom
h.zhou@ecit.qub.ac.uk

Xinbo Gao
Xidian University, Xi'an, China
xbgao.xidian@gmail.com