Command Palette
Search for a command to run...
The Many Faces of Anger: A Multicultural Video Dataset of Negative Emotions in the Wild (MFA-Wild)
Roya Javadi Angelica Lim
Abstract
The portrayal of negative emotions such as anger can vary widely between cultures and contexts, depending on the acceptability of expressing full-blown emotions rather than suppression to maintain harmony. The majority of emotional datasets collect data under the broad label ``anger", but social signals can range from annoyed, contemptuous, angry, furious, hateful, and more. In this work, we curated the first in-the-wild multicultural video dataset of emotions, and deeply explored anger-related emotional expressions by asking culture-fluent annotators to label the videos with 6 labels and 13 emojis in a multi-label framework. We provide a baseline multi-label classifier on our dataset, and show how emojis can be effectively used as a language-agnostic tool for annotation.
Code Repositories
Benchmarks
| Benchmark | Methodology | Metrics |
|---|---|---|
| emotion-classification-on-mfa | MLKNN | F-F1 score (Comb.): 0.34 F-F1 score (NA): 0.42 F-F1 score (Persian): 0.4 V-F1 score (Comb.): 0.39 V-F1 score (NA): 0.42 V-F1 score (Persian): 0.40 |
| emotion-classification-on-mfa | CC - XGB | F-F1 score (Comb.): 0.33 F-F1 score (NA): 0.42 F-F1 score (Persian): 0.28 V-F1 score (Comb.): 0.36 V-F1 score (NA): 0.4 V-F1 score (Persian): 0.33 |
Build AI with AI
From idea to launch — accelerate your AI development with free AI co-coding, out-of-the-box environment and best price of GPUs.