CALLAS

The CALLAS Framework provides a unifying model for multimodal emotion recognition, able to integrate multimodal input data, by fusing them at a higher semantic level, and also involving user and context-aware analysis [see the poster,  a video of the Framework Demonstrator here]

The architecture implements the concept of adaptivity to guarantee independence from the modalities, the devices, the users and the domain/environment, providing the “glue” to easily connect, integrate and apply components. An intuitive graphic interface, the CALLAS Authoring Tool (CAT),  supports application designers: components can be imported (from the SHELF or additional ones), associated to a "state" and "transitions", events and actions. The resulting configuration file is sent to the Framework for execution.

Recently released on  , CALLAS Framework and CAT offer easy design and integration patterns to the open community of developers' for exploitation of the project findings.
Suggested reading:

• An Open Source Integrated Framework for Rapid Prototyping of Multimodal Affective Applications in Digital Entertainment: Abstract

A 1^st Demonstrator  was initially provided to tune the respondence of the Framework to the requirements of the Shelf and Showcases. An evaluation  survey was launched on this screencast on  the C³ to   extend the discussion on the Framework conceptual architecture model for functional relationships between the different Shelf components and their semantic fusion and interpretation.