• 0 Posts
  • 125 Comments
Joined 11 months ago
cake
Cake day: August 8th, 2023

help-circle

  • How good is good do you say?

    We got a pretty good results with CER at 4% and WER at 15%!

    This was on a limited dataset used to test and train which most likely means that if you introduced an even larger dataset with greater variations in handwriting style for testing the numbers might be even worse.

    Very simplified: A risk of a character wrong every 20th character and a word wrong every 7th word. The SER was around 20%.

    There’s an reason why no one has released a good model for western letters yet and why companies pay up to 1€ for capturing data from 10 handwritten pages.

    It will come but OCR isn’t as sexy as developing text2image solutions.





  • To train an AI to recognize handwriting you need a huge dataset of handwriting examples. That is millions of samples of handwritten text + information about what the written text says in every example).

    This is why the best engines only exists as a service in the cloud. The OCR engines you can install lovely that are acceptable, but far from perfect, are commercial. Parascript FormXtra is one of the better commercial ones.

    The only OCR Engine that’s free and really good is Tesseract OCR but it doesn’t handle handwritten text.









  • Yeah, that was you continuing to show how inexperienced you are.

    For a remote exploit to work the computer or device has to expose ports to the network your computer is connected to.

    “Remote” means that the vulnerability does not require local access. So if your friend connects his infected device to your wifi, all devices connected to the same network essentially are at risk, depending on what’s listening on the devices and what vulnerabilities they have.

    Your idea about avoiding bad websites is ridiculous. History is full of examples where third party ads had been created to infect one way or another. That’s ads that users on legitimate site were exposed to. That’s just one little example. There have been numerous examples of malicious sleeping JavaScript code that suddenly wakes up and contacts it’s command-and-control server and then download malicious JavaScript code to unknowing site visitors.

    Furthermore, you didn’t understand my question. Of course antivirus is able to stop malware it recognizes that enters through a remote exploit. The user with antivirus would at least have a chance of knowing that something was up each time and attempt to infect was made.

    You on the other hand would sit there clueless with your little zombie computer and laugh at all them script kiddies.

    But hey… You just continue trying to infect others around you with bad security advice and have a good day. I’m outta here.