The Challenges of Wearable Computing: Part 1
IEEE Micro
Sparse bayesian learning and the relevance vector machine
The Journal of Machine Learning Research
Multimodal human-computer interaction: A survey
Computer Vision and Image Understanding
Speech Emotion Analysis in Noisy Real-World Environment
ICPR '10 Proceedings of the 2010 20th International Conference on Pattern Recognition
Hi-index | 0.00 |
Human-machine interaction could be enhanced by providing information about the user's state, allowing for automated adaption of the system. Such context-aware system, however, should be able to deal with spontaneous and subtle user behavior. The artificial intelligence behind such systems, hence, also needs to deal with spontaneous behavior data for training as well as evaluation. Although harder to collect and annotate, spontaneous behavior data are preferable to posed as they are representative of real world behavior. Towards this end, we have designed a distributed testbed for multisensory signals acquisition while facilitating spontaneous interactions. We recorded audio-visual as well as physiological signals from 6 pairs of subjects while they were playing a bluffing dice game against each other. In this paper, we introduce the collected database and provide our preliminary results of bluff detection based on spatio-temporal face image signal analysis.