As time paces, people are becoming more health conscious. Paying attention to what they eat, counting calories, tracking their daily steps, and whatnot all prove their dedication towards overall well-being. With the advent of artificial intelligence, it was only a matter of time for it to seep into the food and health industry as well.

As per a recent study by Sky News, the researchers at the University of Waterloo in Canada are developing an AI model that can help you track your calories. Once launched, the technology can be a game changer for people who invest their time reading the ingredient labels and manually calculating their calorie intake.

  • Carote 4 Pieces Pots And Pans Set Nonstick, Kitche...

    ₹11,995
    Buy Now
  • The Indus Valley Pre-Seasoned Iron Tawa for Dosa/C...

    ₹714₹1,165
    39% off
    Buy Now
  • Smart Dinning Unbreakable Melamine Bowl Set of 3 P...

    ₹599₹1,999
    70% off
    Buy Now
  • Bar Box 4-Piece Cocktail Shaker Set - Food Grade S...

    ₹11,995
    Buy Now

The report suggests that the current algorithms that estimate the portions of food expect the users to take multiple pictures of their food. The researchers say that the results with this technology would have an error as the picture might not clearly include all the food items on a plate. For example, in a top-down picture of a curry-based dish, you might not be able to get the ingredients inside the bowl.

The researchers thus are integrating language models like ChatGPT. The Sky News report suggests, as per Yuhao Chen from the University of Waterloo, “We're shifting towards using those large language models like ChatGPT [...] to understand what is in the food or maybe ask a basic question [like] 'is this chicken?' "A lot of time, especially for people eating at home, the dish may not be a named dish. It may be just whatever is available in the fridge that they've mixed together."

So, while the existing systems have to study a picture, the upcoming AI system will be able to monitor every spoonful that you eat. Their reports suggest that the system would be able to calculate the volume of food that you eat with a minimal 4.4% error margin. Apart from spoons, Their system would also work for chopsticks and forks. Furthermore, in the later stages of development, the model will be trained to identify a wide range of food items.