'Translate Object' is an iOS app written in Swift that utilizes Core Machine Learning and the Resnet50 Convolutional Neural Network model to identify objects in the mobile device's live camera feed in real-time. The name of the object is then translated to a foreign language of the user's choice in real-time via a Google Cloud Services API. The user can switch languages at any time while using the app. The native language is set to the user's mobile device preference.
This application was created as my 'Hackathon' project at Fullstack Academy. It won the 'Nerd Cred' award for most technically challenging project.
After attempting to create this app using React Native twice, I came to the conclusion that there were certain performance limitations associated with bridging in React Native that could not facilitate the smooth, snappy user experience I wanted. With 4 days remaining until the project was due, I did some research and decided to give myself a crash course in Swift and Xcode. I watched several tutorials and tore though the documentation. On the morning I was scheduled to give my presentation, I finally had a working protype on my iPhone.. at 5:34 am after no sleep what-so-ever... I screen recorded my app in action and uploaded the video just in time to show my classmates and teachers what I had created. I have since continued to fine tune the app, learning more about training Machine learning models, Swift, and mobile app architecture.
NOTE: In order for the translation functionality of this app to work, you will need to enter your own Google Translation API key in the GoogleTranslate.Swift file where it says 'ENTER YOUR API KEY HERE'. (I have hidden my API key for obvious reasons)