The tool read Will Smith's expressions
A computer scientist from Brazil has developed a real-time emotion detection model to evaluate the face of Hollywood actor Will Smith during his speech at the Oscars after he slapped comedian Chris Rock. The expressions are evaluated based on a wheel of emotions. Rondinelli Morais, artificial intelligence specialist and big data, read Will Smith's reactions and the video is now doing rounds on social media.
The tool reads faces from a video and detects feelings based on something like an emotion wheel. There are 4 directions on the wheel: Exciting, Positive, Calming, Negative, and Neutral. The parameters included anger, sadness, fear, boredom and disgust in the negative emotions, whereas in the positive ones they were: emotion, surprise, happiness and relaxation.
Watch the video here:
Will Smith stormed the Oscars stage and struck comedian Chris Rock across the face for joking about his wife. Rock, presenting the best documentary prize with a short comedy routine, had cracked a joke comparing Jada Pinkett Smith's tightly cropped hair to Demi Moore's appearance in the film "G.I. Jane" and suggesting she appear in a sequel.
In a moment that triggered awkward silence and confusion in the Dolby Theatre, Smith strode up to Rock and slapped him, before returning to his seat alongside Jada and shouting profanities.
"Keep my wife's name out of your f*****g mouth," yelled Smith, forcing the producers to bleep out several seconds of audio from the televised broadcast in the United States.