Foodvisor automatically tracks what you eat using deep learning

Meet Foodvisor, a startup that has built a mobile app that helps you log everything you eat in order to lose weight, follow a diet or get healthier. You can add data by capturing a photo of your plate before you eat.

“We’ve spent a little over two years doing research and development before we launched the app in 2018 in France,” co-founder and CMO Aurore Tran told me. Foodvisor has raised $1.5 million so far (€1.4 million).

The company is using deep learning to enable image recognition to detect what you’re about to eat. In addition to identifying the type of food, the app tries to estimate the weight of each item.

Foodvisor tries to evaluate the distance between your plate and your phone using camera autofocus data. It then calculates the area of each food item. The company then tries to extrapolate the volume of each item depending on the type of food.

And of course, if Foodvisor got something wrong, you can manually correct it before you log your meal. Many people give up on nutrition trackers because it’s too demanding. Foodvisor’s technology is all about making the data entry process as seamless as possible.

After that, you get a list of nutrition facts about what you just ate — calories, proteins, carbs, fats, fibers, etc. You can then set a goal, log activities and monitor your progress over time.

The startup has managed to attract 1.8 million app downloads already. It is available on iOS and Android in French, English, German and Spanish. “We have adjusted our product, we’ve enriched our database to better target the American market,” Tran said.

It offers a premium subscription for $5 to $10 per month. In addition to more analysis and diet plans, the main feature of the premium plan is that you can chat with a registered dietitian/nutritionist directly in the app. It turns out that artificial intelligence can’t replace real human nutritionists altogether.