Google’s MI can Categorize Ramen by Shop

Google+ Pinterest LinkedIn Tumblr +

Google’s machine learning software can now identify which store and menu bowls of ramen belong to, according to a new article published on April 2, 2018.

Computers are now well equipped to identify the exact shop a menu item came from, out of 41 seemingly identical bowls of ramen from the same restaurant franchise.

Google’s AutoML Vision was used by a group of researchers to classify every menu item from Ramen Jiro, a Tokyo-based chain of ramen shops. 1,170 photos from each of the 41 shops, were fed as dataset into the software. AutoML took 24 hours to finish training the data, which was further able to decipher the shop that ramen came from with a 95 percent accuracy.

It was initially hypothesized that the model was looking at the color and shape of the bowl or table in the photo, however, it was later found that the software was able to identify specific ramen shops even from photos with the same bowl and table design. The model is accurately designed to identify and distinguish between cuts of meat and the placement of the toppings.

Cloud AutoML software introduced by Google in January 2018, enables developers to create machine learning software through a simplified, drag-and-drop process. They aimed at simplifying AI coding, and training custom vision models through an image recognition tool. Brands such as Urban Outfitters and Disney are already using Cloud AutoML technology to improve the e-commerce shopping experience.


About Author

Has his widespread expertise across the chemicals, food, materials, and energy industries. His enthusiasm and challenging nature, has held him at the forefront of several conferences and seminars. His well-informed knowledge regarding several advancements in such fields, makes him a proficient reporter. Contact Author

Comments are closed.