News for hackers. Stuff that matters.

liabru

Re:How Google Translate squeezes deep learning onto a phone on Wednesday July 29, 2015 @09:18PM (#9972722)
Could you link to it please?

I've not tried it on anything else, but I remember thinking that it has a lot of potential uses. Also I only used it on gray-scale features, but I'm sure it could make use of full RGB too. I'll have to try it some time!

Re:How Google Translate squeezes deep learning onto a phone on Wednesday July 29, 2015 @04:04PM (#9971221)
Hmm looks like they have, well here's another link to it: https://dl.dropboxusercontent.com/u/1672291/scrabble-referee...
Re:How Google Translate squeezes deep learning onto a phone on Wednesday July 29, 2015 @02:23PM (#9970554)
This is great. I particularly like that they also automatically generated dirty versions for their training set, because that's exactly what I ended up doing for my dissertation project (a computer vision system [1] that automatically referees Scrabble boards). I also used dictionary analysis and the classifier's own confusion matrix to boost its accuracy.

If you're also interested in real time OCR like this, I did a write up [2] of the approach that worked well for my project. It only needed to recognize Scrabble fonts, but it could be extended to more fonts by using more training examples.

[1] http://brm.io/kwyjibo/

[2] http://brm.io/real-time-ocr/

Nice work!

A little trick you can use here is squared distance which avoids the relatively expensive square root (also do the square in-place rather than pow) so your distance function becomes:

(d - a) * (d - a) + (e - b) * (e - b) + (f - c) * (f - c)

You will need to also square your threshold values.

Re:How to create 2D visibility/shadow effects for your game on Tuesday March 11, 2014 @07:12PM (#7382847)
Nice example, but ray casting like this is overkill for this effect and with a low number of rays it's pretty inaccurate.

I prefer using the technique shown in this classic article on 2D soft shadows: http://archive.gamedev.net/archive/reference/programming/fea...

The basic idea is to project the geometry, rather than ray cast. I've implemented this sort of technique in the past and it works well.