Beacon Transcript

Information which Matters to You

Wednesday, January 27, 2021
Log in
  • National News
  • National News
  • Business
  • Business
  • Technology
  • Technology
  • Health
  • Health
  • Science
  • Science
  • About Us
    • Contact Us
    • Authors & Contributors
    • Privacy Policy GDPR
  • About Us
    • Contact Us
    • Authors & Contributors
    • Privacy Policy GDPR

Recent Articles

Ancient Babylonian Clay Tablet Holds the First Trigonometric Table

Ancient Babylonian Clay Tablet Holds the First Trigonometric Table

August 26, 2017 By Clayton Meason

The CDC Draws Attention To The Spike In Cyclosporiasis Cases

The CDC Draws Attention To The Spike In Cyclosporiasis Cases

August 11, 2017 By Troy G. Bennett

Travis Kalanick Resigns from His Position as Uber’s CEO

Travis Kalanick Resigns from His Position as Uber’s CEO

June 22, 2017 By Troy G. Bennett

Amazon Refunds Parents Whose Children Made Purchases Without Permission

Amazon Refunds Parents Whose Children Made Purchases Without Permission

June 2, 2017 By Clayton Meason

McDonald’s Has Been Quietly Altering Its Vanilla Ice Cream Recipe

McDonald’s Has Been Quietly Altering Its Vanilla Ice Cream Recipe

May 21, 2017 By Jennifer Licata

Washington Post Will Be Expanding Its Newsroom

Washington Post Will Be Expanding Its Newsroom

December 28, 2016 By Troy G. Bennett

PlaNet Uses Pixels for Geolocation

February 28, 2016 By Helen Bradford

Email, RSS Follow
"New Deep-Learning Engine Can Recognize Location with Uncanny Accuracy"

The sofware was able to identify 48 percent of the times the continent from which the pictures were takes.

BEACON TRANSCRIPT – Recently, a team of computer engineers from Google has announced a computer application that is capableof recognizing where a picture was taken without using geotags or a large database. PlaNet uses pixels for geolocation, and according to its makers, it can outwrestle even the savviest travelers.

Forget about bulky computer algorithms and about geolocation tags, because the new software developed in Google’s computer labs will be able to tell you exactly where you took your vacation picture just by analyzing the background.

When devising the new of the deep-learning engine, the scientists though that it is both redundant and resource-consuming for a machine to memorize each detail from a picture. Instead, the team sought to create an algorithm that can identify the location by analyzing its background.

The team explained that PlaNet can do just that. The deep-learning software can divide a whole picture into pixels. As we known from the 101 computer class, each pixel is different, kind of like human fingerprints. The software is trained to detect these slight variations in size and location and to compare them.

For instance, if the system is fed with pictures of Paris, it will immediately begin to search for similar pixels. If more than 1000 pictures depict the Eiffel Tower, the machine will learn the size and location of the pixel grid associated with the Eiffel Tower. This means that if someone uploads the 1001 picture of Eiffel Tower, the machine will immediately say that the picture was taken in Paris, France which is situated in Western Europe.

Now, you are probably wondering how the team managed to make the computer identify the photo’s location without geotags? It was no easy task, to be sure. Luckily, Google’s has been experimenting for some time with deep-learning engines, so it was just a matter of transformation.

The team fed the computer with over 120 million pictures taken from Flickr. After they’ve uploaded the pictures, the team tested the computer’s capability of geotagging on a sample of 90 million pictures.

Instead of trying to memorize all the pictures, the deep-learning algorithm began to dissect all the pictures into pixels and to memorize the configuration of each one. To see if the software learned anything, the team decided to put a little strain on the newly-born application.

Fed with a sample of 30 million picture and a human opponent, the app’s goal as to recognize as many pictures as possible. As it happens, PlaNet was able to win 28 out of 50 rounds against the human opponent.

Moreover, PlaNet, which acts up like a neural network, managed to identify landmarks at street level 3.6 percent of the time and big landmarks 10.3 percent of the time. The accuracy went up by 48 percent when the app was asked to identify in what part of the globe the picture was taken.

Although the results seem pretty low, PlaNet actually managed to beat its human adversary when it comes to median proximity. When asked to identify the location, the human opponent was off by 1.441 miles, while the app was off by only 703 miles.

PlaNet uses pixels for geolocation and the app is so small in size (336 MB) that it can be installed on virtually any smartphone.

Photo credits:www.pixabay.com

Email, RSS Follow

Filed Under: Science Tagged With: deep-learning engine, geolocation, geotags, google, identify location, landmarks, pixel grid, PlaNet

Recent Posts

  • Security Guard List As Suspect in 1974 Stanford University Chapel Slaying Kills Himself June 29, 2018
  • Texas Teens Arrested After Decomposing Body of Elderly Woman Found in Garage June 28, 2018
  • Montana Woman in Custody After Forcing Ex to Have Sex with Her June 27, 2018
  • Newly Released Audio Transcript Proves Monalisa Perez Objected to Deadly YouTube Prank June 26, 2018
  • Virginia Woman Shoots New Zealand Man Who Smashed Glass Door with Brick   June 26, 2018
  • Arizona Woman Lied About Kidnapping to Cover Boyfriend’s Slaying June 25, 2018
  • Brooklyn Man Arrested in Niagara Falls Allegedly Chopped off Pregnant Wife’s Arms with Steak Knife June 25, 2018

Related Articles

  • jupiter with the biggest moon of jupiter, ganymede, in front, a satellite, and three moons

    Ganymede, the Biggest Moon of Jupiter, is Revealing New Data (Study)

    May 8, 2018
  • segment of the bayeux tapestry which includes halley's comet

    Medieval Tapestries and Old Scrolls to Offer Proof of the Existence of Planet Nine?

    May 7, 2018
  • stars and galaxies in the universe

    Large-Scale Map of the Universe Reveals Thousands of Early Galaxies (Study)

    Apr 10, 2018
  • landscape view of the milky way and a space observatory

    The Milky Way Might be Getting Bigger, Believes a New Study

    Apr 5, 2018
  • NASA’s Odyssey Captures New Images of Phobos and Deimos, the Moons of Mars

    Feb 27, 2018
  • old book of shakespeare plays open to a page with his image

    Software Reveals Possible Inspiration for 11 Shakespeare Plays

    Feb 14, 2018
  • praying mantis on black background

    Praying Mantis and 3D Glasses Help Discover A New Form of Vision

    Feb 12, 2018
  • two men sitting squatting and looking at the oman desert

    Mars Simulation Tests, Conducted in the Oman Desert

    Feb 9, 2018
  • indian stone tools on grey background

    Ancient Indian Stone Tools Reveal New Things about Humanity (Study)

    Feb 5, 2018
  • Naked mole rats.

    Naked Mole Rats Are Not Affected by Aging, New Study Reveals

    Jan 29, 2018

Categories

  • Business
  • Entertainment
  • Health
  • National News
  • Nature
  • Science
  • Stocks
  • Technology
  • US
  • World

Copyright © 2021 beacontranscript.com

About · Privacy Policy · Terms of Use · Contact

This website uses cookies to ensure you get the best experience on our website. Learn more.