Highly precise and accurate frame by frame video emotion analysis
This app allows you to analyze faces in videos, frame by frame, and download detailed CSV or JSON files reports to get super accurate emotional data based on scientific models.
All you need to do is to upload a video from your device and start the analysis. Our Emotion AI will analyze the main face in the video, frame by frame, to provide you with precise data.
With this tool you will be able to detect:
- Arousal
- Valence
- 98 affects (Russel, Sherer, Klaus model)
- 38 affects (Paltoglou and Thelwall model)
- quadrants (Russel, Sherer, Klaus model)
- 6 emotions (Ekman model)
- Attention
- Number of faces
Once you finish the analysis, you will be able to download a JSON or CSV file with the data detected by our AI.
Pricing:
The cost is calculated based on the number of analysis per month and the average view time.
The minimum charge is one minute per view.
Check our price calculator!
Scientific references:
- Russell, J., Lewicka, M. & Niit, T. (1989). A cross-cultural study of a circumplex model of affect. Journal of personality and social psychology, 57, 848–856.
- Ekman, P. (1999). Basic emotions. Handbook of cognition and emotion, 98(45-60), 16.
- Scherer, Klaus. (2005). Scherer KR. What are emotions? And how can they be measured? Soc Sci Inf 44: 695-729. Social Science Information. 44. 695-792.
- G. Paltoglou and M. Thelwall (2013), “Seeing Stars of Valence and Arousal in Blog Posts,” in IEEE Transactions on Affective Computing, vol. 4, no. 1, pp. 116-123, Jan.-March 2013, doi: 10.1109/T-AFFC.2012.36.