This app converts your Apple iOS device into a Image Recognizer Demo app.
The user clicks the Image Recognizer Demo app icon then they are taken right into the app. They are greeted with a two top options for importing an image for use with this app. On the left the user can click the camera icon to use the built in camera to take a phone then submit it to the app for object translation. Then on the upper right the user can click the 'library' button to be taken to the devices internal library and use those images for translation. In this app we are using images from the library only on the iPhone X simulator app. In the top middle of the app is the app's title "Image Recognizer". This is just a label only.
In the middle of the screen is where the image selected will be show. When the app starts in the bottom there is a label propting the user to 'Select an Image' to be imported. Once imported when the image is translated the answer will show up on this label to what the recognizer app believes it is seeing. As you can see there are times when its correct and other times where its incorrect. Plus working with this app i've noticed that sometimes changing the angle of a device brings up the correct answer too.
The Inception v3 Core ML Model detects the dominant objects present in an image from a set of 1000 categories such as trees, animals, food, vehicles, people and more.
Screen caps taken from App running on an iPhone X device running iOS 11.0. Coded in Swift 4.