Hacker News Reader

This is really neat!

I uploaded an image of a gopher and it automatically detected the shape and drew it on the code.

http://assets.avi.io/screen-shot-2016-11-15-f4w73.png

This is impressive. Would love insights on how you did that.

I think it's simple, every character is a pixel, you just have to change the color of the characters to match the same pixel on the uploaded image.

It's a great idea btw.

Agreed

It is not so simple. The characters are not the same colour as in the image. Look at the eyes and the nose for example. It seems to be doing a more advanced kind of image feature detection.

It is, just the white pixels on the image are black text on the poster, and the same with black pixels, so Gopher's eyes (and the other black parts) disappear.

There is a Light Text option, with that white pixels are grey so you can see the eyes :)

https://s18.postimg.org/4jg01zwux/gopherposter.jpg

Reply on news.ycombinator.com