Weighted Accuracy (WA): 79.4Unweighted Accuracy (UA): 79.8Weighted F1: 79.4
Papers with Code · CC BY-SA 4.0
Emotion Recognition
Emotion Recognition is an important area of research to enable effective human-computer interaction. Human emotions can be detected using speech signal, facial expressions, body language, and electroencephalography (EEG). <span class="description-source">Source: Using Deep Autoencoders for Facial Expression Recognition </span>
논문
EmoA: 77.4EmoV: 61.9
EmoA: 72.1EmoV: 61.7
Concordance correlation coefficient (CCC): 0.638
Accuracy: 96.10
Accuracy: 40.34
Accuracy: 86.70%
CCC (Valence): 0.648CCC (Arousal): 0.764
1'": 1
Accuracy: 90.2F1: 0.891
Top-3 Accuracy (%): 14.73
CCC: 0.377
Unweighted Accuracy: 65.20
Accuracy: 74.8
F1-score (Weighted): 51.17
Top-3 Accuracy (%): 13.73
5-class test accuracy: 66.13%
UA: 0.89WA: 0.53Macro F1: 0.54
Unweighted Accuracy (UA): 0.429
WAR: 83.61%
Accuracy: 80.08%
CCC: 0.6466753
Accuracy: 81.58%
CCC: 0.676