University of Cambridge researchers have built an AI-powered robot chef that learned how to make eight salad recipes by watching a video of a human making them. Grzegorz Sochacki, a PhD candidate, and his colleagues filmed eight simple salad recipes and then used a public neural network to train their robot chef.
This neural network was already programmed to identify a range of different objects, including the fruits and vegetables used in the salad recipes. The robot then analyzed each frame of video using computer vision techniques and identified various objects as well as features such as a knife or ingredients. Correctly identifying the ingredients and the actions of the human chef enabled the robot to determine which one of the recipes was being prepared.
- EVERYTHING YOU NEED AT A GLANCE - With a 15.6” Full HD (1080p) smart display and Fire TV, family organization and entertainment will look brilliant....
- YOUR NEW KITCHEN TV - With Fire TV, you can stream over 1 million movies and TV episodes. Watch favorites with subscriptions to Netflix, Hulu, Prime...
- ALEXA VOICE REMOTE FOR FIRE TV - Easily browse and navigate Fire TV content from across the room. Quickly access popular Fire TV streaming apps with...
It’s amazing how much nuance the robot was able to detect. These recipes aren’t complex – they’re essentially chopped fruits and vegetables, but it was really effective at recognising, for example, that two chopped apples and two chopped carrots is the same recipe as three chopped apples and three chopped carrots,” said Grzegorz Sochacki, a PhD candidate from Cambridge’s Department of Engineering, the paper’s first author.
