Re:Machine Learning with TensorFlow Enabled Mobile Proof-Of-...
by davidkuhta on Tuesday October 17, 2017 @11:36AM
(#15491804)
Also worth mentioning:
- Fast: Needed a one-second average processing time (product-code iamge -> OCR pipeline)
- Accurate: Goal was to achieve 95% string recognition accuracy (with improvement via active learning)
tl;dr:
- Customised text recognition of codes printed on goods
- UX flow designed to (i) help users correct errors, and (ii) gather additional labelled images