11-18-2024, 08:11 PM
I really think this handwriting recognition is going to go somewhere.
The future is sure to have it!
https://youtu.be/evMilwVBHAQ?t=1089
One I saw that was really cool was automated check scanning. Old lisp derivative called Lush. Written by Yann LeCun who is as per his page Chief AI Scientist, Meta. The quote about the checks was:
a handwriting recognition system used by many banks across the world to read checks automatically. In fact, some ATM machines made by NCR (that can read checks) run compiled SN code on embedded DSP boards. A version of this check reader written in SN is embedded in large check reading engines sold by NCR and OrboGraph. According to some estimates, this system automatically reads between 10 and 20 percent of all the checks written in the US.
not shabby for a DSP embed-able recognition engine...
Wonder how it compares to resource utilization vs a TensorFlow GPU derived solution.
The future is sure to have it!
https://youtu.be/evMilwVBHAQ?t=1089
One I saw that was really cool was automated check scanning. Old lisp derivative called Lush. Written by Yann LeCun who is as per his page Chief AI Scientist, Meta. The quote about the checks was:
a handwriting recognition system used by many banks across the world to read checks automatically. In fact, some ATM machines made by NCR (that can read checks) run compiled SN code on embedded DSP boards. A version of this check reader written in SN is embedded in large check reading engines sold by NCR and OrboGraph. According to some estimates, this system automatically reads between 10 and 20 percent of all the checks written in the US.
not shabby for a DSP embed-able recognition engine...
Wonder how it compares to resource utilization vs a TensorFlow GPU derived solution.