The company is relying on AI and machine learning to replace the background in videos without the need for a green screen, and the interface looks as easy as slapping on a Snapchat or Instagram filter onto a video. How it all works is explained at the source link, there are a lot of technicalities but its all worth it cause the results looks pretty convincing. As of now, YouTube is running a beta version of the technology with only a handful of the content creators. Will it only be available for Google Pixel 2 devices, provided they come with a special Pixel Visual Core chip that applies intelligent processing for images? So, it shouldn't be a surprise the rollout of the new feature for mobile video.
"Our new segmentation technology allows creators to replace and modify the background, effortlessly increasing videos' production value without specialised equipment", reads a blog post from Google Research software engineers, Valentin Bazarevsky and Andrei Tkachenka.
Furthermore, the specific segmentation task to compute a binary mask separating foreground from the background for every input frame (three channels, RGB) of the video was created.
Separating the background is an impressive feature in itself, but Google took it forward, by making a program to run on limited hardware on a smartphone rather than a desktop computer.
To watch videos on YouTube Go app, you need to complete few steps such as choosing language, adding phone number and email id.
Google's artificial intelligence researchers have a new trick up their sleeves: giving selfie videos a new background the way you can with still photos today.
Google's new tech, however, does not mean that the end of the green screen is anyway near. As a result, the AI's background replacement quality not quite on par with a green screen, but Google's technique does not require editing skills or extra work. Although the roll-out started last month, the option to download YouTube videos didn't appear in more than several countries.