The diet app LoseIt! released a new deep learning feature called Snap It that lets users take photos of their food and then it automatically logs the calorie count and nutritional information.
Using the NVIDIA DIGITS deep learning training system on four TITAN X GPUs, the company trained their network on a vast database of 230,000 food images and more than 4 billion foods logged by Lose It! users since 2008.
“The more people use this, the more it improves,” said Edward W. Lowe, data scientist at Lose It. “The goal is to get the accuracy high enough in six months so it won’t even need to ask you for validation.”
“Without the GPUs, we never would have initiated this project,” Lowe said.
Read more >
Lose It!’s New Food Recognition App Counts Calories
Oct 04, 2016
Discuss (0)
AI-Generated Summary
- LoseIt! has introduced a new feature called Snap It that uses deep learning to log calorie counts and nutritional information from photos of food.
- The feature was developed using NVIDIA's DIGITS deep learning training system on four TITAN X GPUs and trained on 230,000 food images and over 4 billion foods logged by users since 2008.
- Edward W. Lowe, data scientist at Lose It!, stated that the accuracy of Snap It will improve with more user data and aims to achieve high accuracy within six months.
AI-generated content may summarize information incompletely. Verify important information. Learn more
About the Authors
