1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

What's Happening in This Picture?

Discussion in 'BlackHat Lounge' started by The Scarlet Pimp, May 26, 2015.

  1. The Scarlet Pimp

    The Scarlet Pimp Senior Member

    Joined:
    Apr 2, 2008
    Messages:
    876
    Likes Received:
    3,309
    Occupation:
    Chair moistener.
    Location:
    Cyberspace
    here's something for you fans of "captcha solvers" and "image identifiers" to play with... :batman:

    ---

    What's Happening in This Picture?

    It's a question that human minds can answer instantly in most cases. Show me a photo of nine uniformed men standing on a diamond-shaped field and I'll tell you, "It's a baseball game."

    Computers have a much harder time answering such questions. But this new technology just might be a breakthrough. Read on to learn more and try it for yourself...

    What is Wolfram Image ID?

    First, Stephen Wolfram gave us , a "computational knowledge engine" that provides computed answers to natural-language questions instead of merely Web pages that contain some of the words in the questions. Then he developed the Wolfram Language, a programming language that draws much of its power from the algorithms and databases underpinning Wolfram|Alpha.

    Now, Stephen has added a simple function to his language, ImageIdentify, that answers the question, "What's going on in this picture?"

    Let's go back to that example of a baseball game. Make it nine little kids instead of grown men, and I'll refine my answer to, "Little League game."

    Such instant identifications are second nature to humans; few of us realize what vast stores of facts and billions of lightning-fast computations are required to pull of this simple feat.

    Artificial-intelligence experts know that image-recognition is incredibly complex, and it's going to be utterly necessary in the future of self-driving vehicles, self-piloting drones, and other autonomous machines. Image-recognition is a problem that people like Stephen Wolfram are very keen to solve.

    The ImageIdentify function is a simple way for a programmer to invoke a very complex program that operates on Wolfram|Alpha's database of several trillion facts. This program has been exposed to over 10,000 "training images" and their associated identities or tags, e. g., "coffee mug."

    Given an image as input via the ImageIdentify function, the program returns the tag of the image that is the closest fit to the input image.

    You can try the ImageIdentify function on any image you like, at the Wolfram Language Image Identification Project Web site. You'll also be helping the function expand and refine its knowledge base.

    You can drag an image into the project's input area, or upload an image file from your hard drive. The project will take just a few seconds to analyze the image and return a tag. Then you'll be asked to rate the project's performance on a scale labeled "Great," "Could be better," "Missed the point," or "What the heck?!"

    If your rating is less than "great," the project will offer second guesses and ask you to tell it what you think the tag should be.

    ...and All I Got Was This Lousy T-Shirt

    Finally, you can leave your email address to be notified when ImageIdentify learns more about "your kind of image" and, more importantly, to be entered in a daily random drawing for an "I taught ImageIdentify" T-shirt. That ought to get you a few "What the hecks?" of your own!

    My own tests produced very uneven but not unamusing results. ImageIdentify correctly recognized a flower and even gave me its genus. It deemed me a "person." An origami fish was a "vertebrate."

    ImageIdentify guessed correctly that one photo of a baseball stadium was a "ballpark" but thought that another one was a casino.

    The FAQ attributes some bloopers to "training errors." For example, ImageIdentify may have been trained to associate the word "Chihuahua" with several examples of the dog, one of which was wearing a party hat. So if you upload a picture of a party hat, it may be identified as a "Chihuahua." Wolfram and company are working on such things.

    There are other image-recognition Web sites to which you can compare Wolfram's ImageIdentify. Their results are generally similar but can be amusing in their own ways.

    Examples include CamFind, Clarifai, MetaMind, Orbeus, and AlchemyAPI.

    https://www.imageidentify.com/

    http://cloudsightapi.com/api

    http://www.clarifai.com/#demo

    https://www.metamind.io/vision/general

    https://rekognition.com/demo/concept

    http://www.alchemyapi.com/products/demo/alchemyvision
     
    • Thanks Thanks x 2
  2. archon10

    archon10 BANNED BANNED

    Joined:
    Oct 10, 2011
    Messages:
    1,181
    Likes Received:
    1,668
    I wanna see what it says for porn.

    "Woman sucking dick."
    "Anal"
    "Gang bang."
     
    • Thanks Thanks x 1
  3. Capo Dei Capi

    Capo Dei Capi BANNED BANNED

    Joined:
    Oct 23, 2014
    Messages:
    754
    Likes Received:
    1,734
    IBM also has one

    http://visual-recognition-demo.mybluemix.net/
     
  4. sohom

    sohom Senior Member

    Joined:
    May 26, 2013
    Messages:
    990
    Likes Received:
    175
    Location:
    not in Past
    Good
    Coder can use them for bypassing human verification :p
     
  5. HoNeYBiRD

    HoNeYBiRD Jr. VIP Jr. VIP

    Joined:
    May 1, 2009
    Messages:
    7,171
    Likes Received:
    8,157
    Gender:
    Male
    Occupation:
    Geographer, Tourism Manager
    Location:
    Ghosted
    i'm seconding this, google's no captcha recaptcha needs to be cracked once and for all
     
  6. bmminc

    bmminc Regular Member

    Joined:
    Nov 29, 2014
    Messages:
    391
    Likes Received:
    284
    it kinda sucks. guessed a $50 bill as a "Card case"

    Porn it always says "person"

    put a simple picture of an arrow going up on a graph and it said "device".

    Not good.
     
    • Thanks Thanks x 1
  7. tony_d

    tony_d Elite Member

    Joined:
    Jun 22, 2013
    Messages:
    2,582
    Likes Received:
    3,176
    Location:
    1600 Amphitheatre Parkway, Mountain View CA
    Indeed.

    For the below picture, it guessed 'dining room', then asked whether 'chain saw' would have been better when I said it had missed the point...
    wtf.

    [​IMG]
     
  8. PaulDante

    PaulDante Newbie

    Joined:
    Jun 12, 2013
    Messages:
    11
    Likes Received:
    3
    It's like they didn't see Terminator. THE MACHINES ARE AFTER US!
     
  9. Ambitious12

    Ambitious12 Elite Member

    Joined:
    Jun 26, 2014
    Messages:
    3,096
    Likes Received:
    609
    Occupation:
    No Occupation
    Location:
    Among the Stars