Exciting developments in the field of emotion recognition!
Our team dedicated to the neuroscientific area has been working on capturing the information related to emotional states encoded in variation of physiological signals such as heart rate, skin conductance, and electroencephalogram (EEG).
One trick we employ is to transform electrophysiological signals into images (such as the Gramian Angular Field), making it easier to visualize and analyze the data. The timeseries-to-image conversion enables the usage of image-based architectures, and in particular Convolutional Neural Networks, for understanding the physiological data.
We are excited about the potential applications of this technology, from healthcare to education and beyond. By better understanding the emotional states encoded in physiological signals, we can help improve people’s quality of life.