HEROES

HEROES is a new video database for emotion recognition. In HEROES, 16 non-professional actors express four emotions (happiness, interest, disgust and boredom). The videos have been acquired by using four GoPro cameras to record whole body movements in two different scenarios: observational and interaction with another person. Feature vectors that describe the joints movement are included, together with emotion annotation by experts to be used as benchmark for further processing.

Click on the links below to download:

• Read me file with instructions
• The HEROES dataset (size: 480MB): it is composed by 256 mp4 videos. It is divided into four folders: Boredom, Happiness, Interest and Disgust. Each emotion contains 16 folders related to the actors, and for each actor there are four videos: two related to Scenario A, two related to Scenario B.
• The feature vectors (size: 880kB)
• The coordinates of each joint (size: 307kB)
• The expert annotations

If you are using these resources please cite: I. Mannocchi, K. Lamichhane, M. Carli, F. Battisti, “HEROES: A Video-Based Human Emotion Recognition Database”, 10^th European Workshop on Visual Information Processing, 11-14 September, 2022. 

The investigators involved in this research are:

• Ilaria Mannocchi
• Kamal Lamichhane
• Federica Battisti
• Marco Carli