Call for papers: Affect, Emotion and Behavior Processing in Human-Machine Interaction
A Special Issue of APSIPA Transactions on Signal and Information Processing
Brief description:
Affect, emotion and behavior processing is a budding research area with a major focus on a more natural interaction between humans and machines. The advancements in affective computing from multimodal cues can be expected to lead to a major breakthrough in human-machine interaction in the near future. There are a wide range of issues and challenges in this area, particularly in modelling, recognition and synthesis of affective, mental and social behaviors and phenomena, as represented by speech, text, facial expressions, gaze, image, video, and physiological signals (e.g., heart rate, brain waves, skin conductance and respiration).
Currently, the capacity of model-based affective and emotional communication from multimodal cues is still too limited to provide a completely natural human-machine interaction. Although it is possible for multimodal systems to be able to communicate meaning concretely through language and facial expression, their ability to recognize and synthesize affect, emotion and behavior information is quite primitive. A significant aspect of natural interaction includes the ability to understand the affect and emotion information, to adapt audio-visual generation to the characteristic of particular users, and to fulfil the goal of communication.
Thus, affect, emotion and behavior processing raises new issues for computational science. It involves,
- How to represent emotions, affective and behavior states in a computational manner on the basis of psychological models?
- How to design the architecture of systems to model and process the related data using these concepts?
- How to work with appropriate machine learning techniques to deal with data to train general models or models in an end-to-end manner for affective computing?
- Whether other key areas of computational research areas, such as evolutionary learning algorithms and deep neural networks, reinforcement learning, speech/visual/physiological signals recognition and synthesis can provide essential tools to address the significant challenges of affective computing?
This special issue aims to collect together a global and comprehensive overview of the current state of this field. It will present updated results in research frontiers, algorithm development and its applications in automatic affect, emotion and behavior representation, recognition, and synthesis in human-machine interaction.
Potential Topics interests but are not limited to:
- Theories of emotions from psychology and their applications to computer science
- Models of emotion and physiological information (e.g., heart, brain, thermal signals, etc)
- Automatic emotion recognition & synthesis from physiological signals, facial expression, head movements and gestures, body postures and gestures, and speech
- Expressive text-to-speech and expressive language translation
- Sentiment analysis and opinion score
- Affect and Emotional Conversation Generation
- Smart environments & digital spaces (e.g., in a car, or digital artworks)
- Clinical and biomedical applications (e.g., autism, depression, pain etc.)
- Emotion mining in text, images, videos, film, multimedia data
- (Multimodal) naturalistic data sets and annotations
- End-to-end emotion classification and synthesis from multimodal cues
- Emotion analytics applications
- Perceptual evaluation of virtually expressive agents and robots
- Affective interaction with virtual agents and robots
- Multimodal behavior signal analysis and interpretation
Editor in Chief APSIPA T-SIP
Tatsuya Kawahara, Professor, School of Informatics, Kyoto University, JAPAN
Guest Editor(s) of the special issue:
- Haizhou Li, National University of Singapore
- Rafael E. Banchs, Institute for Infocomm Research, Singapore
- Dong-Yan HUANG, Institute for Infocomm Research, Singapore
- Ming Li, Duke Kunshan University, China
Schedule of submission and publication:
Submission Deadline: Saturday, 30 June 2018
Publication Date: December 2018
Papers are published upon acceptance, regardless of the Special Issue publication date.