Skip to main content
. 2024 May 28;24(11):3484. doi: 10.3390/s24113484

Table 4.

Mostly used emotion recognition datasets.

Expression Database Posed/ In the Wild Images/Videos Type Subjects Expressions References
Facial
Macro
FER-2013 [60] in the wild 35,000 images gray - anger, disgust, fear, happiness, sadness, surprise, neutral [61,62,63]
CK+ [64] posed 593 images mostly gray 123 neutral, sadness, surprise, happiness, fear, anger, contempt, disgust [61,65,66,67,68,69]
JAFFE [70] posed 213 images gray 10 neutral, sadness, surprise, happiness, fear, anger, disgust [61,67,68,71]
RAF-DB [72] in the wild 29,000 images color 300+ anger, disgust, fear, happiness, sadness, surprise, neutral [72]
AffectNET [73] in the wild 950,000 color - neutral, happiness, sadness, surprise, fear, disgust, anger, and contempt [45,74]
Aff-Wild2 [75] in the wild 2,800,000 color 458 neutral, happiness, sadness, surprise, fear, disgust, anger + valence–arousal + action units 1, 2, 4, 6, 12, 15, 20, 25 [76]
Facial
Micro
CASME [77] posed 195 images color 19 amusement, sadness, disgust, surprise, contempt, fear, repression, tense [78]
CASME II [79] posed 247 images color 26 happiness, disgust, surprise, repression, and others [40,41,78,80]
Gestures MHHRI [81] posed 48 videos color 18 self-/acquaintance-assessed personality, self-reported engagement [12]
Static
Poses
EMOTIC [82] in the wild 18,313 images color 18 anger, fatigue, fear, happiness, sadness, pain, confidence, and others [13]