To add the Watson Visual Recognition service, you need to go to the IBM Cloud Dashboard and select Watson Services, then Browse Services, where you can then find and select the Visual Recognition service:
Next, from the Visual Recognition page (which is shown in the following screenshot), you can choose a location and resource group for the service instance and then click on Create to actually create the instance of the service that you can use in your project:
Once you have created the Visual Recognition service instance, it will be listed on your cloud dashboard:
After creating the service instance, you should be able to create what is referred to as Service Credentials (or the API key) by clicking on New Credentials or View Credentials in the Service credentials section.
The following screenshot shows the Service credentials page:
We will need to refer to this API key in the next few sections of this chapter, so keep it handy!
Now we are ready to get going with the actual Visual Recognition project. In this example (a version is available on GitHub), the Watson Visual Recognition service is used to perform object detection using the TensorFlow object detection API.