Diverse Reactions of Engagement and Attention Mind States (DREAMS) Dataset, is a collection of user videos and engagement and attention self-labels. The videos were captured while the users were watching a set of educational and humorous stimuli. The purpose of this dataset is to enable researchers to predict users' engagement and attention states and uncover the relationship between the two. 

Dataset Statistics:

Total number of subjects: 32(13 Females, 19 Males)

Total number of videos: 781

Duration of each video: ~40 seconds

Total Duration: ~521 minutes

Train: 600 

Test: 181

Dataset Class Distribution:

Highly Engaged/Attentive: 260

Highly Engaged/Not Attentive: 2

Engaged/Attentive: 220

Engaged/Not Attentive: 59

Barely Engaged/Attentive: 32

Barely Engaged/Not Attentive: 133

Not Engaged/Attentive: 8

Not Engaged/Not Attentive: 67

Sample Frames