What it does is reproduce an image with a set of words which may or may not be associated with it.
Check out this example of my photo represented with a bunch of JS keywords/operators:
One can choose a preset example or upload his own image or use camera as input source (if supported).
The available controls for now are:
- source - select between image you can upload or video stream from camera (if supported). The more contrasty the source is, the better
- grid size - adjusts how detailed the output will be
- color - controls the hue of output
- shades - controls the quantity of color shades
- background color - in case you'll feel creative
- font - needs to be a valid css font-family value
- vocabulary - space-separated words to be used for the output. The more words of different lengths, the better. Note: as there will likely be a lot of small "square" areas on your source image, make sure you have short enough words (like "yo", or "lol", or ";-)") in your vocabulary that can suit those areas. If you don't provide any, "??" will be used instead.
All images used for example presets belong to their respective owners.