AI, Artificial Intelligence, also called machine intelligence which is the intelligence manifested by machines. Hence, AI is simply the capability of a machine to imitate intelligent human behavior. AI has become so much advanced these days that it even helps people in their day to day activities such as searching photos, breaking down language barriers in Google Translate, typing emails on the go, or getting things done with the Google Assistant and so on.
Now talking about the Google AI, it’s presenting its Move Mirror AI experiment which is a special unique experiment that creates the flipbook-like experience and lets us explore our images in our browser by following our real-time moves and matches them with more than 80 thousands of images of people doing similar poses. For this, you only need your web browser and a webcam. You just need to move or give poses in front of your webcam and then move mirror matches your pose with the similar poses of thousands of people in this world. It’s so much fun using Move Mirror AI.
Google said, Move Mirror helps to show computer vision techniques like pose estimation, the ability to detect humans and their poses from image or video data, to all the people and can be available to anyone using a computer with a webcam. Apart from making a video only, it also helps to create a GIF of the Move Mirror AI and share it with friends.
To build this experiment, Google uses the PostNet neural network model, a special software that uses pose estimation even the picture resolution power is very low, blurry or black and white. This PostNet is powered by Tensorflow.js, a library for training and deploying ML models in the browser and on Node.js