Live Object Detection Using Google ML Kit
In this project, we built live object detection on the mobile application. We used flutter dart language. This project idea is very unique in AI/ML Projects because we used Google ML Kit Packages or Libraries and our project application name is "Detector Objects".
Google ML Kit is a mobile SDK that brings Google's machine learning expertise to Android and iOS apps in a powerful yet easy-to-use package. In This Project, We Learnt How To detect objects and After Detecting How to fit the objects on a square alignment its a very tuff part of our project part because without detection and without identification this project was not meaningful or useful. We first tried TensorFlow libraries but flutter removed this library a year ago, so we have many difficulties building this project after some research or surfing we finally got a solution and implemented this project on use a google ml kit.
we used this plugin in our project,
- google_mlkit_object_detection:
- camera: With, the help of the pub dev website. This website gives all flutter plugins and packages.
Also, We used a splash screen in our application. whenever the users open our application first shows a splash screen after the users redirect to the live object detection page. this splash screen logo we created in the Figma application. Figma is the best application for building logos, ui/ux screens, and many more.
We faced many challenges when we start the projected camera running well but object detection issues and many errors occurred. so we find so many solutions in google or other search sites but not more solutions we find out because all solutions see used TensorFlow. After We try our logic and we got a solution to our error. This Error is object detection but can't see a square around an object so all detections are mixed up and no one understands what this app actually used.
Built this project we learned new concepts and learned how to use new libraries, also most of the AI/ML projects are built in python and used TensorFlow, and pandas libraries. but we decided to try something new and unique. so we developed this project on the mobile application as an android project.
So, that's it for our project and in the future, we try to build more projects and participate in more hackathons...
Built With
- android-studio
- camera
- dart
- figma
- flutter
- googlemlkit