Skip to content

QualityPoint Technologies News

Emerging Technologies News

Menu
  • About Us
  • Technology
  • Medical
  • Robots
  • Artificial Intelligence (AI)
  • 3D Printing
  • Contact Us
Menu

Huge Discount Offer: 14 ebooks + 2 courses

Background-swap technology suitable for mobile phones announced

Posted on March 5, 2018

Video creators: Want to swap backgrounds? Knock yourselves out.

Google researchers have been working on a way to let you swap out your video backgrounds using a neural network—no green screen required.

It’s rolling out to YouTube Stories on mobile in a limited fashion, said TechCrunch.

Neural network video segmentation in YouTube stories.

Video segmentation is a widely used technique that enables movie directors and video content creators to separate the foreground of a scene from the background, and treat them as two different visual layers.

Video content creators know that a scene’s background can be separated from the background treated as two different layers. The maneuver is done to achieve a mood, or insert a fun location or punch up the impact of the message.

The original frame (left) is separated in its three color channels and concatenated with the previous mask (middle). This is used as input to neural network to predict the mask for the current frame (right).

The operation, said the two on the Google Research site, is “a time-consuming manual process (e.g. an artist rotoscoping every frame) or requires a studio environment with a green screen for real-time background removal (a technique referred to as chroma keying).”

Google researchers know how much people like to trick others into thinking they’re on the moon, or that it’s night instead of day, and other fun shenanigans only possible if you happen to be in a movie studio in front of a green screen. So they did what any good 2018 coder would do: build a neural network that lets you do it.

An example image from dataset carefully annotated with nine labels – foreground elements are overlaid over the image.

This “video segmentation” tool, as they call it (well, everyone does) is rolling out to YouTube Stories on mobile in a limited fashion starting now — if you see the option, congratulations, you’re a beta tester.

“Google is developing an artificial intelligence alternative that works in real time, from a smartphone camera,” Grigonis wrote.

Real-time video segmentation in action.

Google designed a network architecture and training procedure suitable for mobile phones focusing on the following requirements and constraints:

  • A mobile solution should be lightweight and run at least 10-30 times faster than existing state-of-the-art photo segmentation models. For real time inference, such a model needs to provide results at 30 frames per second.
  • A video model should leverage temporal redundancy (neighboring frames look similar) and exhibit temporal consistency (neighboring results should be similar)
  • High quality segmentation results require high quality annotations.

This is great news for a lot of folks — removing or replacing a background is a great tool to have in your toolbox and this makes it quite easy. And hopefully it won’t kill your battery.

News Source: https://research.googleblog.com/2018/03/mobile-real-time-video-segmentation.html

Related Videos:

MIT’s Deep-learning system generates Videos that predict what will happen next in a scene | QPT

Nest Cam Outdoor – Mystery Solved. Watch many Crazy Funny Video Clips captured by Nest Cam.

MIT’s “CornerCameras” system can detect objects or people in a Hidden Scene

NVIDIA’s AI produces Fake Human Photos with Unbelievable Quality | QPT

Loom.ai – This New Technology can generate your realistic 3D Avatar from a single selfie | QPT

AI Dancer – LuminAI watches Dancing Person and improvises its own moves using AI | QPT

Share

Related News:

  1. Now anyone can explore machine learning, no coding required
  2. Google Pixel Buds to translate 40 languages in real time
  3. AI System “IDA” helps Farmers to monitor Cows
  4. Google has launched a website for showing Semantic Experiences
Master RAG ⭐ Rajamanickam.com ⭐ Bundle Offer ⭐ Merch ⭐ AI Course

  • Bundle Offer
  • Hire AI Developer

Latest News

  • ​Firebase Studio: Google’s New Platform for Building AI-Powered Applications April 11, 2025
  • MIT Researchers Develop Framework to Enhance LLMs in Complex Planning April 7, 2025
  • MIT and NVIDIA Unveil HART: A Breakthrough in AI Image Generation March 25, 2025
  • Can LLMs Truly Understand Time Series Anomalies? March 18, 2025
  • Can AI tell us if those Zoom calls are flowing smoothly? March 11, 2025
  • New AI Agent, Manus, Emerges to Bridge the Gap Between Conception and Execution March 10, 2025
  • OpenAI Unveils GPT-4.5, Promising Enhanced AI Performance February 28, 2025
  • Anthropic Launches Claude Code to Revolutionize Developer Productivity February 25, 2025
  • Google Unveils Revolutionary AI Co-Scientist! February 24, 2025
  • Microsoft’s Majorana 1 Chip: Revolutionizing Quantum Computing with Topological Core Architecture February 20, 2025

Pages

  • About Us
  • Basics of 3D Printing
  • Key Innovations
  • Know about Graphene
  • Privacy Policy
  • Shop
  • Contact Us

Archives

Developed by QualityPoint Technologies (QPT)

QPT Products | eBook | Privacy

Timesheet | Calendar Generator

©2025 QualityPoint Technologies News | Design: Newspaperly WordPress Theme