I am implementing firebase's Recognize text in images in my Android App. I can successfully get the result from the API by block/line/element.
https://firebase.google.com/docs/ml-kit/android/recognize-text
My question is, in the result data package, when I use getConfidence()
in block/line/element, all return "NULL". However, when I test the API in the Cloud Vision API page and in the API demo page, there is confident return in the JSON.
Anyone know if there is way to get the confidence of the block/line/element
For on-device Text Recognition we - sadly - do not provide confidence numbers at this moment. As part of an upcoming update we may reconsider this. FYI, the challenge is that it's not trivial to maintain calibration of confidence levels between updates to the underlying models.