It implements the work in the EmotiW18 challenge. In the engagement intensity prediction task, we win the champion and the details are refered in our paper "Deep Recurrent Multi-instance Learning with Spatio-temporal Features for Engagement Intensity Prediction. International Conference on Multimodal Interaction (ICMI’18), ACM."
We only provide the code for feature extraction, training and testing. As our deep multi-instance model is still modified for further improvement, we will release the model code soon.