New computer program first to recognize sketches more accurately than a human

Jul 30, 2015

by Science Daily

Researchers from Queen Mary University of London (QMUL) have built the first computer program that can recognise hand-drawn sketches better than humans.

Known as Sketch-a-Net, the program is capable of correctly identifying the subject of sketches 74.9 per cent of the time compared to humans that only managed a success rate of 73.1 per cent.

As sketching becomes more relevant with the increase in the use of touchscreens, the development could provide a foundation for new ways to interact with computers.

Read the full article by clicking the name of the source located below.

2 comments on “New computer program first to recognize sketches more accurately than a human

  • 1
    Light Wave says:

    For real ??? What human did you test ? ? A rubbish one ? And could this computer programme decipher my sketches of abstract notions…I bet it couldn’t…humans are perfectly capable of recognising sketches in fact humans are more than adequate why do we even want a programme like this…

    Report abuse

  • Presumably a real human. And they don’t know everything. I’m sure almost everyone in europe can tell the difference between a pigeon and seagull, but what about people in south America who have never seen a seagull or a pigeon?

    How many can correctly identify two WWII aircraft? How about different types of sailing rigs? What about people who rush though the test and don’t take time to look at every picture properly? Or those who do and (if it is a timed test) don’t reach the last few? Humans can make all sorts of mistakes, and that pushes the score down to ~70%. A computer can spend more time per picture than a human can and finish faster, and have access to neigh unlimited knowledge. Or at least google and google image search.

    And I’m guessing that several million eventually rising to one or two billion would like program like this worldwide.

    Report abuse

Leave a Reply

View our comment policy.