CMOSE: Comprehensive Multi-Modality Online Student Engagement Dataset with High-Quality Labels
Project page is here
Video clip name
Each video clip is named as videoX_Y_personZ, which means it is the Yth clip of the Zth subject from coaching session X.
We extract the second level features from OpenFace. The extracted files are stored under "secondfeature/videoX_Y_personZ.csv". These features include:
Gaze Direction and… See the full description on the dataset page:
https://huggingface.co/datasets/zhangchengpu/CMOSE.
