Google cloud AutoML Vision doubt

29 views

Skip to first unread message

Manuel James

unread,

Oct 16, 2018, 10:33:14 AM10/16/18

to Google App Engine

hello everyone, i will keep it simple because its a very generic question but i don't have any answer yet, the problem its that in my university a company set a little "Investigation group" and its related to aumented reality and we need a image identificaction prograrm so i read only that google cloud service, most acurrated cloud atml vision. the real question is... not the cost or the dificulty, its the libert we have whit the AI tool, the most concerned i am is whit you can have a respond whit it? like, i scan a mug image and it send me to a page that i whant, for something to say to netflix for example.

PS:I am really sorry if this "Post" its not welcome here ( i mean, if it not belong here, if not please directme to the right one, i only need a simple answer)

A lot of thanks beforehand

George (Cloud Platform Support)

unread,

Oct 16, 2018, 1:50:03 PM10/16/18

to Google App Engine

Hello Manuel,

Your core concern, formulated as: "the libert we have whit the AI tool, the most concerned i am is whit you can have a respond whit it?", is not directly understandable at first look. Assuming you would like to trigger some specific reactions based on results from an AutoML Natural Language call: this is possible, based on custom labels. In fact, AutoML Natural Language allows you to train custom machine learning models, with your own custom labels. These labels may be detected and further used in your program to direct you to a page you want, let's say Netflix. The "AutoML Natural Language Beginner's guide" page provides you with related detail.