Today’s Startup of the Day is SnapCalorie.

SnapCalorie takes a photo of any meal and generates an accurate calorie count. The app leverages advanced computer vision and depth sensors to produce results that are 2x better than the average human nutritionist. SnapCalorie focuses on the root of successful AI models: training data. They created a unique dataset of about 5000 images of food—individually weighing each ingredient and adding complicating factors like oils, “mystery sauces”, and more. In tandem with depth sensors on supported devices, SnapCalorie is able to produce more accurate caloric reads than their competitors.

SnapCalorie was founded by Wade Norris and Scott Baron. The founding team has a rich background in computer vision: in sum, they previously co-founded Google Lens, Cloud Vision API, and published a CVPR paper demonstrating the first algorithm to outperform a professional nutritionist at calorie counting. Their foray into the nutrition space aims to unite cutting edge computer vision techniques with the blooming interest in what we put in our bodies—to enable us to better understand what we’re really consuming, without making inaccurate judgements and wasting time on our phones while at the dinner table.