Re:How Google Translate squeezes deep learning onto a phone
by joe_the_user on Wednesday July 29, 2015 @03:20PM
(#9970964)
It seems your dissertation paper is behind something password protected [1]. It would be nice to see that too.
Can't get [1]https://www.dcs.shef.ac.uk/intranet/teaching/campus/projects...
Re:How Google Translate squeezes deep learning onto a phone
by liabru 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
by JabavuAdams on Wednesday July 29, 2015 @03:12PM
(#9970890)
Glad there's prior art on that. I had a small project where I iterated all the fonts on the system and used them to generate glyph training images. The next step was to dirty them up, but I never continued the project.
More generally, I really like the idea of generating controlled synthetic images and then messing them up for regularization.
Re:How Google Translate squeezes deep learning onto a phone
by megalodon on Wednesday July 29, 2015 @07:02PM
(#9972194)
Funny, just read an article today proposing the same feature detection algorithm (the one you called 'grid merge'). Have you tried applying these techniques on scanned/photographed documents?
Re:How Google Translate squeezes deep learning onto a phone
by liabru 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
by megalodon on Wednesday July 29, 2015 @10:23PM
(#9972895)
"We also investigated hierarchical features where the image is overlaid with a grid of cell size c × c and pixels withins each cell are added up. This is same as downsampling the image and using the raw pixels in the downsampled image as features." (p. 3)
http://www.eecs.berkeley.edu/Pubs/TechRpts/2009/EECS-2009-15...
Re:How Google Translate squeezes deep learning onto a phone
by pedrosorio on Thursday July 30, 2015 @12:37AM
(#9973190)
Sounds similar to one level of a pyramid:
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/