1
$\begingroup$

I have been completely unable to find details on the underlying model that is used for the Palm Pilot Graffiti handwriting recognition. Given the computing power available for handheld PDA technology, I have always been quite surprised this was possible at that time. I was hoping for a source with information about the model, e.g., features, recognition method, etc.

$\endgroup$
  • $\begingroup$ I'm not sure how useful it is, but the Palm system was "stolen" from Xerox's Unistrokes system, which was written up here. $\endgroup$ – Peter K. Jul 7 '17 at 20:33
5
$\begingroup$

If you are looking for a good explanation of some of the methods used in projects like GRAIL look here:

Back to the Future of Hand Writing Recognition

$\endgroup$
  • $\begingroup$ That may have been the best tutorial I have ever gone through. After a single pass I feel vaguely confident that I could implement a simple detector on a microcontroller. It is an unbelievably simple approach to a seemingly very complicated problem. Thanks. $\endgroup$ – adfriedman Jul 8 '17 at 8:35
  • $\begingroup$ This one is a great catch! +1. $\endgroup$ – Royi Jul 8 '17 at 9:29
3
$\begingroup$

Surprised? DARPA (GRAIL project?, et.al.) supported handwriting recognition research circa 3 decades earlier, and on mainframes less powerful than a PalmPilot 1000.

Hawkins talks a bit about implementing Graffiti in a Computer History Museum oral history.

$\endgroup$
  • $\begingroup$ Your link for the oral history is bad. $\endgroup$ – Yet Another User Jul 7 '17 at 23:12
  • $\begingroup$ seems to be, yeah. $\endgroup$ – Yet Another User Jul 12 '17 at 2:05
2
$\begingroup$

It seems most of the data is in Graffiti (Palm OS) - Wikipedia.
It seems the technology is related to Unistrokes technology (U.S. Patent 5,596,656, granted in 1997).

Since the model relies on single stroke and using symbols that are exceptionally well separated from each other graphically I would be assume parameterizing the stroke would be a great way to get well separated features. Just the coordinate and the relation between them will be perfect.

$\endgroup$
  • $\begingroup$ I had looked this up on Wikipedia before I asked the question, but it wasn't clear to me whether the patent dispute was over a unistroke input method or the actual recognition system design. I figured that it was the former as the patent was repeatedly disputed. $\endgroup$ – adfriedman Jul 8 '17 at 8:30
  • $\begingroup$ Do you know if the recognition system was the same? $\endgroup$ – adfriedman Jul 8 '17 at 8:37
  • $\begingroup$ Well, if they made the effort to only support "Single Stroke" which leaves very little to discern between the hypothesis. $\endgroup$ – Royi Jul 8 '17 at 9:30
1
$\begingroup$

if you can get into the IEEE Explore Library

Pull up

D. J. Burr, "Designing a Handwriting Reader," in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. PAMI-5, no. 5, pp. 554-559, Sept. 1983.

and then look at the citations, which include both the Palm and Xerox Patents. Xerox claimed that Palm infringed so you might pick up some clues.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.