Google Vision/automl Models

[Tshego the beauty maker Monoge]

Hi,

I am currently using google vision for object detection. Unfortunately I cannot see the models that are used. 

1.) Is it possible to view the models used by google vision

2.) Can I upload new models i.e yolo, mobilenet v2, inception etc? If so, how can that be accomplished.

The only option i see is to re-train a model. But I want to use models that are already trained.

Scenario.

I have a local application that runs tiny yolo. I send the image for analysis only when my local app found/detected something. The problem is that the google vision is not more accurate(returns false positives) than the yolo  I am running locally.

Thank you.

[Steeve]

Hi,

To answer your questions:

Q.  Is it possible to view the models used by google vision

 A.  No

Q.  Can I upload new models i.e yolo, mobilenet v2, inception etc? If so, how can that be accomplished.

A.  In order to use your own model, you need to use the Cloud AutoML Vision which is currently in beta release. I'd recommend to start using the quickstart to understand how it works

Thank you.