I got the inspiration when I was working on a previous team project that was an Arduino project. We were needed that the robot should identify that object itself. So, I got an idea to make a live camera app that detects every object so that we can identify what to do next .! I know this idea has been already implemented, this is new to me. But I am confident about the future that I'll definitely find a new idea and will sustained it.

What it does

It's an android application that detects every object using ML computer vision implementation. It identifies and bitmap every object present in the image and labels it accordingly.

How I built it

I build it by implementing TensorFlow TFlite. It provides a pre-trained python module of computer vision that detects every object.

Challenges I ran into..!

By implementing Tfline I was very confused about how to access the TFlite file. Task Library is a cross-platform library that makes it easy to deploy TensorFlow Lite models with just a few lines of code in my apps. It was totally new to me. I don't have the idea about that.

Accomplishments that we're proud of..!

I don't think I have done any accomplishments. I don't think my project gonna finish on time. But I'll submit this project whatever I have done on this project.

What I learned.!

Literally, I learned lots of new things. Using the TFlite module in any app is totally new to me. Making a sequential code to read the data from the user data and finding the objects is really cool.

What's next for Labeling Image Using Machine Learning TFlite..!

After labeling the Image, I'll work on the live camera labeling using computer vision. This is really challenging for me but I know I'll definitely find a way to conqueror my all problems Because, till I never try I won't succeed it.

Built With

  • android
  • computervision
  • kotlin
  • tensorflow
Share this project: