in ,

FORGET SEARCH KEYWORDS — THIS NEW MACHINE LEARNING BASED SYSTEM LETS YOU SEARCH WITH DRAWN SKETCHES!

Search via a drawn sketch with the help of Machine Learning Frameworks like Tensorflow or Torch.

DigitalTrends.com has written an interessting article about a new machine learning based system: Draw a Sketch and find the corresponding eCommerce results in a shop ( [Link: www.digitaltrends.com] ).

Photo-based search lets you draw of an item and then find that particular product – the developer hopes within six months you’ll be able to log into a shopping site and use this technology.

1. What is the idea?

Researchers of the Queen Mary University of London’s (QMUL) School of Electronic Engineering and Computer Science ( [Link: www.eecs.qmul.ac.uk] ) have created and trained a Deep Learning ( [Link: en.wikipedia.org] ) framework to recognize hand-drawn sketches and use them to find real-life products on a shopping website.

Their network was trained to connect sketches to photos based on a comprehensive data set of about 30 000 sketch-photo comparisons. Like a human brain the software has learned to understand the connection between a hand drawing and a matching eCommerce product in a shop – it is very accurate and helps to speed up search results.

2. to start to create it on your own?

Do some research based on CNN (Convolutional Neural Networks) – [Link: deeplearning.net][Link: ufldl.stanford.edu][Link: www.tensorflow.org] .

3. Want to create a Business based on Machine Learning?

We tell you what to do first – Contact us: [Link: www.onetopp.com]

Image: ai.icymi.email/tag/sketch-based-search/

Tags: MACHINE LEARNING, SKETCH BASED SEARCH, ARTIFICIAL INTELLIGENCE, ,

What do you think?

0 points
Upvote Downvote

Total votes: 0

Upvotes: 0

Upvotes percentage: 0.000000%

Downvotes: 0

Downvotes percentage: 0.000000%

Schreibe einen Kommentar

Deine E-Mail-Adresse wird nicht veröffentlicht. Erforderliche Felder sind mit * markiert.

GOOGLES TRANSPORT PROTOCOL QUIC WILL SPEED UP THE INTERNET TODAY: FASTER GAMES, VIDEO STREAMING AND REAL TIME VIDEO MESSAGING.

BLOCKCHAIN-BASED VIDEO STREAMING OPEN SOURCE APPLICATION MIGHT BE THE BITTORRENT SUCCESSOR!