Datasets

 

Facial Expression Database

The MUG Facial Expression Database was created by the MUG group. It contains several image sequences of an adequate number of subjects for the development and the evaluation of facial expression recognition systems that use posed and induced expressions. The image sequences were captured in a laboratory environment with high resolution and no occlusions. In the first part of the database eighty six subjects performed the six basic expressions according to the ’emotion prototypes’ as defined in the Investigator’s Guide in the FACS manual. The image sequences start and stop at neutral state and follow the onset, apex, offset temporal pattern. The publicly available sequences count up to 1462. Landmark point annotation of 1222 images is available as well. In the second part of the database eighty two of the subjects were recorded while they were watching an emotion inducing video. This way several authentic expressions are recorded.