9 ms·

This looks cool but I wish it showed examples of different kinds of handwriting being detected. Cursive? European numbers vs American numbers (the 1 especiall

This looks cool but I wish it showed examples of different kinds of handwriting being detected. Cursive? European numbers vs American numbers (the 1 especially)? Doctor (or other badly rendered hand writing)? Seems like this technology isn't really all that useful if it doesn't work on various inputs, especially 'unclean' or 'sloppy' inputs.

mlaskin 9y ago
Thanks for the comment! The algorithm works on a large variety of handwriting (currently only support English). It's a NN that was trained on ~100k different handwriting examples, and it's pretty robust to cursive / sloppy handwriting. Haven't tried European vs American numbers yet but I'll definitely give it a try now that you've piqued my curiosity
- ItendToDisagree 9y ago
  Thanks for the reply! The European 1 often looks like an upward facing triangle without a bottom line. Surprisingly frequently it has caused European's in the US to have issues with the IRS/Taxes because the European 1 is often interpreted by Americans as a 7 (because Americans do not use a bar across the center of the 7 or Z). A household size of 7 instead of 1 can be a real back breaker!
- 6stringmerc 9y ago
  Pretty cool project / release! How does it do with bubbly teenage girl script? Example: http://www.annakoren.com/images/love7.gif http://www.annakoren.com/images/love7.gif Best to ya.
  - mlaskin 9y ago
    thanks for the question! It handles that kind of script pretty well, mainly because we had a good amount of training examples from college student handwriting
mootothemax 9y ago
>European numbers vs American numbers (the 1 especially)? Honestly I think 9 will be the bigger issue, to my eyes the European version is near identical to lower-case G.
- literallycancer 9y ago
  You mean q? Because g has that loop at the end.
  - mootothemax 9y ago
    No, I mean G precisely because it has that loop at the bottom. I see it near universally in Poland, and 100% in Germany but with a terribly small sample size.
  - IanCal 9y ago
    Both, really. Check out this small sample from MNIST https://www.researchgate.net/profile/Amaury_Lendasse/publication/264273647/figure/fig1/AS:295970354024489@1447576239974/Fig-18-0-9-Sample-digits-of-MNIST-handwritten-digit-database.png https://www.researchgate.net/profile/Amaury_Lendasse/publica...

mlaskin 9y ago

Thanks for the comment! The algorithm works on a large variety of handwriting (currently only support English). It's a NN that was trained on ~100k different handwriting examples, and it's pretty robust to cursive / sloppy handwriting. Haven't tried European vs American numbers yet but I'll definitely give it a try now that you've piqued my curiosity

ItendToDisagree 9y ago
Thanks for the reply! The European 1 often looks like an upward facing triangle without a bottom line. Surprisingly frequently it has caused European's in the US to have issues with the IRS/Taxes because the European 1 is often interpreted by Americans as a 7 (because Americans do not use a bar across the center of the 7 or Z). A household size of 7 instead of 1 can be a real back breaker!
6stringmerc 9y ago
Pretty cool project / release! How does it do with bubbly teenage girl script? Example: http://www.annakoren.com/images/love7.gif http://www.annakoren.com/images/love7.gif Best to ya.
- mlaskin 9y ago
  thanks for the question! It handles that kind of script pretty well, mainly because we had a good amount of training examples from college student handwriting

mootothemax 9y ago

>European numbers vs American numbers (the 1 especially)? Honestly I think 9 will be the bigger issue, to my eyes the European version is near identical to lower-case G.

literallycancer 9y ago
You mean q? Because g has that loop at the end.
- mootothemax 9y ago
  No, I mean G precisely because it has that loop at the bottom. I see it near universally in Poland, and 100% in Germany but with a terribly small sample size.
- IanCal 9y ago
  Both, really. Check out this small sample from MNIST https://www.researchgate.net/profile/Amaury_Lendasse/publication/264273647/figure/fig1/AS:295970354024489@1447576239974/Fig-18-0-9-Sample-digits-of-MNIST-handwritten-digit-database.png https://www.researchgate.net/profile/Amaury_Lendasse/publica...