Special Issue description

Although great advances have been obtained in the, so-called, Looking at People field, it is only recently that attention is focused on problems that have to do with more complex and subconscious behaviors. For instance, personality and social behaviors are only starting to be explored from the computer vision and multimedia information processing perspectives. This is because, among other reasons, the lack of data and benchmarks to evaluate this sort of tasks. Nevertheless, the availability of massive amounts of multimodal information together with the dominance of social networks as a fundamental channel where users interact, have attracted the interest of the community in this direction of research. Tools for effectively analyzing these sort of behaviors have a major impact into everyone's life, with applications in health (e.g., support for mental disorders), security (e.g., forensics, preventive applications), human computer/machine/robot interaction (e.g., affective/interactive interfaces) and even entertainment (e.g., user-tailored systems).

This special issue focuses in all aspects of computer vision and pattern recognition devoted to the automatic analysis of human behavior in social media from visual and multimodal information.  The focus is on the analysis of human behaviors that are not visually obvious, i.e. unconscious behaviors and situations in which the sole visual analysis is insufficient to provide a satisfactory solution. Submissions in other aspects of looking at people may be considered as well.


Prospective articles should make fundamental or practical contributions to the field. Topics of interest include, but are not limited to:

  • Human behavior analysis from visual and multimodal information, with emphasis on unconscious behaviors, including, but not limited to: personality analysis, deception detection, social behavior analysis, etc.

  • Human behaviour analysis in the context of social media

  • Multimodal content extraction for human behavior analysis

  • Multimodal machine learning, deep learning, active learning, and transfer learning for human behaviour analysis in social media

  • Multimodal zero-shot learning, and unsupervised learning for the analysis of unconscious human behaviors

  • Crowdsourcing, community contributions, and social multimedia

  • Information fusion techniques for the analysis of human behavior in social media

  • Multimodal information retrieval, categorization and clustering of social media

  • Large-scale and web-scale multimodal analysis of social media

  • Multimodal analysis of human intent and human perception in social media

  • Benchmarks, datasets and evaluation methodologies for unconscious behavior analysis

  • Explainability and fairness in multimodal AI systems for human behavior analysis

  • Applications of unconscious behavior analysis methods, e.g., medicine, sports, commerce, lifelogs, travel, security, environment.


Submission guidelines

All the papers should be full journal length versions and follow the guidelines set out by International Journal of Computer Vision: https://link.springer.com/journal/11263.


Manuscripts should be submitted online at www.editorialmanager.com/visi/ choosing

"S.I. : Analyzing Human Behavior from Social Media Data"

as article type. When uploading your paper, please ensure that your manuscript is marked as being for this special issue. Information about the manuscript (title, full list of authors, corresponding author’s contact, abstract, and keywords) should be also sent to the corresponding editors (see information below).


Submitted papers should present original, unpublished work, relevant to one of the topics of the Special Issue. All submitted papers will be evaluated on the basis of relevance, significance of contribution, technical quality, scholarship, and quality of presentation, by at least three independent reviewers. It is the policy of the journal that no submission, or substantially overlapping submission, be published or be under review at another journal or conference at any time during the review process.


There are no news registered in