1

I'm looking around for different solutions to a problem I'm having and right now I'm looking into the iOS Vision api.

Does anyone know if it's possible to use the text/feature detection methods provided by Vision for custom objects in an image, or custom typefaces?

Andy Jazz
  • 49,178
  • 17
  • 136
  • 220
Axel
  • 463
  • 6
  • 19
  • Have you come to an answer yet? I would also like to detect custom glyphs with Vision Framework, not just alphabet/kanji/etc. – BFeher Apr 17 '18 at 07:10
  • I used TuriCreate to train my own image classifier instead. Worked fine for my purpose, although may not always. – Axel Apr 18 '18 at 13:28
  • Thanks this is what I'm currently doing too; Turi Create trained model in CoreML. Just thought it would make a much cleaner solution to have Vision detect glyphs and then CoreML decipher them instead. – BFeher Apr 18 '18 at 13:41

0 Answers0