FindingEmo: An Image Dataset for Emotion Recognition in the Wild
This dataset addresses the problem of limited naturalistic data for emotion recognition researchers, though it is incremental as it builds on existing datasets by expanding scope.
The authors introduced FindingEmo, a dataset of 25k images for emotion recognition that focuses on complex scenes with multiple people in natural settings, annotated for valence, arousal, and emotion labels.
We introduce FindingEmo, a new image dataset containing annotations for 25k images, specifically tailored to Emotion Recognition. Contrary to existing datasets, it focuses on complex scenes depicting multiple people in various naturalistic, social settings, with images being annotated as a whole, thereby going beyond the traditional focus on faces or single individuals. Annotated dimensions include Valence, Arousal and Emotion label, with annotations gathered using Prolific. Together with the annotations, we release the list of URLs pointing to the original images, as well as all associated source code.