Hi all - Does anybody have any experience in Google Vision API or similar technology for photo tagging?
The use case application - we have lots (10,000+) of photos of objects (think different types of household items), and we want to tag them with text like "hammer" and "dish" and "stove." We have a training set of tagged items, and would like to train a machine to tag the rest. Have played with some Python code using Tensorflow, no real traction just yet.
How would you approach this? Any ideas or referrals to people (academic or commercial) who know more about this than I do are welcome.
-- Carrie
------------------------------
Carrie Beam
Teaching Associate Professor
University of Arkansas
Walnut Creek CA
carrie@carriebeamconsulting.com------------------------------