Personal tools
You are here: Home SIGs Speech SIG IEEE Transactions on Affective Computing Special Issue on Naturalistic Affect Resources for System Building and Evaluation

IEEE Transactions on Affective Computing Special Issue on Naturalistic Affect Resources for System Building and Evaluation

This special issue focuses on the introduction, presentation, and discussion of novel and existing mono- and multimodal affective resources. Alternatively, ways to better exploit existing corpora by improved standardization and combination can be addressed. Necessary steps in this direction comprise mapping schemes to overcome the peculiarities of the field – such as categorical, complex or dimensional, and unstable annotation, and measurements to automatically assess similarity, type, and quality of resources. Also needed are new ways to establish semi-supervised processing of large resources by media tagging, or ways to better bundle efforts of the community, e.g. by shared and distributed collection and annotation of data. Finally, for better exchange and comparability of reported results, partitioning and evaluation strategies will benefit from further discussion. The issues mentioned may be exemplified by novel naturalistic resources or by exploiting existing ones.

CALL FOR PAPERS
IEEE TRANSACTIONS ON AFFECTIVE COMPUTING
Special Issue on Naturalistic Affect Resources for System Building and Evaluation


There has been a constant development of larger and more naturalistic databases in the field of Affective Computing over the last decade – still, the bottleneck remains: many research issues in the field can hardly be addressed due to the evident lack of suited data, and the existing large body of mono-modal data needs addition of multimodal resources stemming from different modalities such as audio, video, physiology, text, etc. – in particular in real-life context and of interaction: emotional behavior is of great importance for social interaction, as emotions serve communicative and social functions and convey information about people’s thoughts and intentions towards the others. In addition, multi-cultural and multilingual data is still considerably sparse. This is even truer when it comes to data in natural or working system contexts.
This special issue focuses on the introduction, presentation, and discussion of novel and existing mono- and multimodal affective resources. Alternatively, ways to better exploit existing corpora by improved standardization and combination can be addressed. Necessary steps in this direction comprise mapping schemes to overcome the peculiarities of the field – such as categorical, complex or dimensional, and unstable annotation, and measurements to automatically assess similarity, type, and quality of resources. Also needed are new ways to establish semi-supervised processing of large resources by media tagging, or ways to better bundle efforts of the community, e.g. by shared and distributed collection and annotation of data. Finally, for better exchange and comparability of reported results, partitioning and evaluation strategies will benefit from further discussion. The issues mentioned may be exemplified by novel naturalistic resources or by exploiting existing ones.
Articles are invited in the area of mono- and multimodal resources for research on emotion and affect. Topics include but are not limited to:
  • Novel corpora of mono- or multimodal affective data in interaction - in particular with high diversity (subjects, language, culture, noise, health state, etc.)
  • Semi-supervised data collection and annotation
  • Measures for quantitative corpus quality, type, and similarity assessment
  • Standardisation and mapping of annotation for inter-corpus exploitation
  • Data for real-life application and system building
  • Long-term recorded databases with rich contextual information
  • Rich and novel annotations and annotation types (e.g. acting, regulation, confidence, etc.)
  • Partitioning and testing protocols for affective resources
  • Evaluations on novel or multiple corpora
  • Community-based and distributed establishment of affective resources

Submissions must not have been previously published, with the exception that substantial extensions of conference papers can be considered. The authors will be required to follow the Author’s Guide for manuscript submission to the IEEE Transactions on Affective Computing at: http://www.computer.org/portal/web/tac/author. Full manuscripts should be submitted electronically through IEEE’s Manuscript Central: https://mc.manuscriptcentral.com/taffc-cs. Be sure to select “Special Issue on Naturalistic Affect Resources” as the Manuscript Type, rather than “Regular Paper.” This will ensure that your paper is directed to the special issue editors. IEEE Tools for Authors are available online at: http://www.ieee.org/organizations/pubs/transactions/information.htm. Inquires can be directed to Kristen Anderson, Allen Press Inc., Administrator for TAC, toac@computer.org (Tel: 800-627-0326 x256 or 785-843-1234 x256).

Schedule:


Submission Deadline: November 1st, 2010
Notification of Acceptance: April 1st, 2011
Final Manuscripts Due: June 1st, 2011

Guest Editors:


Björn Schuller, Technische Universität München, Germany
Ellen Douglas-Cowie, Queen's University Belfast, UK
Anton Batliner, Universität Erlangen-Nürnberg, Germany
Document Actions
Powered by Plone

Portal usage statistics