Object Detection and Tensorflow

An overview of how Python is used for object detection

What is Object Detection?

Like the name suggests, object detection deals with detecting objects in an environment. Specifically, here, we are dealing with detecting certain objects in a robot’s surroundings using cameras. The camera’s visual input is processed to determine whether the object of concern is in the visual field or not. In FTC, we use Java and TensorFlow Lite’s library to work with Object Detection. You can take a look at the page on FTC Object Detection.


TensorFlow is a machine learning framework created by Google Brain that allows us to write code in Python to work with Google-built machine-learning models and create cool applications based on machine learning concepts that Google has already worked on for us. It is a relatively difficult concept to learn, so we don’t recommend you dive straight into this. But, if you are interested, feel free to take a look.

Where to learn

To get started with TensorFlow, you can either look at our page on FTC Object Detection, or go to the TensorFlow Website to learn. There are also great courses out online that you could look into but we recommend only diving into those once you’ve got a good grasp of Python.