Drawing Pictures for a Guessing Game and your Refrigerator

Drawing Pictures for a Guessing Game and your Refrigerator

By Max Candocia


February 24, 2016

Correctly determining what an image is supposed to be is usually easy for a human but often difficult for a computer. I want to make an applet where you can make a drawing and then the computer tries to guess the image correctly. Unfortunately, this requires a bit of training data. If you feel like contributing (or just wasting time drawing something on an html5 canvas), draw something on the canvas below.

For the sake of having a decent number of a few categories , I would try drawing one of these:

cat, dog, horse, banana, rainbow, tree, flower, car, wizard, pirate, ninja, axe

You can change the color you are drawing with by using the palette chooser, change its size using the size scale, and you can choose to erase everything with the erase button (and switch back to marker when you are done). You can also download your picture in another window by clicking "download" so you can print it out and hang it on your refrigerator. It won't close your current window.

If you want to submit your drawing, click "Submit Image", and it will be sent to the server. For security reasons, you will not be able to access the images once they are sent to the server and you leave the page. When you submit the image, you grant me (non-exclusive) permission to use, modify, and distribute the image. I will be programming a neural network for the guessing game as time goes on and post a similar page that features the game.

The palette chooser was made with Spectrum JS, and the canvas drawing was made with canvas.js. I may change the drawing library in the future to have some more sophisticated features.

Brush Size: 5

What is this a picture of?


Recommended Articles

Subreddits That Get You the Most Awards

Which Subreddits are most likely to generate awards for their users?

Sketching and Guessing with Keras

Draw a picture and see if the computer can guess it. You can also submit images for the algorithm to develop more possible results.