BP4D+ Multimodal Spontaneous Emotion Database

 Well-annotated, multimodal, multidimensional spontaneous emotion corpus of 140 participants. 

 Background: 

 In the last 10 years, research on facial expression analysis has shifted its focus from posed behavior to non-posed (i.e., spontaneous) behavior. This shift has increased the difficulty of such analyses, but also their ecological validity and practical utility. 

 Technology Overview:  

 We present a well-annotated, multimodal, multidimensional spontaneous emotion corpus of 140 participants. Data were acquired from a variety of facial sensors that included high-resolution 3D dynamic imaging, high-resolution 2D video, thermal (infrared) sensing, and contact physiological sensors that included electrical conductivity of the skin, respiration, blood pressure, and heart rate.   

http://binghamton.technologypublisher.com/files/sites/rb440.jpg

https://www.pexels.com/photo/collage-of-portraits-of-cheerful-woman-3807758/

 Advantages:  

• Emotion inductions highly varied.
• Facial expression annotated for both the occurrence and intensity of facial action units from 2D video by experts in the Facial Action Coding System (FACS).
• The database includes derived features from 3D, 2D, and IR (infrared) sensors, and baseline results for facial expression and action unit detection.
• Database further includes a diverse array of participants.