Skip to content
KozhaAkhmet edited this page Mar 7, 2024 · 7 revisions

Welcome to the depth-car wiki!

About the Project

Depth Car is an autonomous car with Monocular depth estimation using MiDAS, TensorFlow Lite and OpenCV on The Raspberry Pi 4, written in C++.

This project uses Midas Depth Estimation as a core part of the project. So it is recommended to have basic knowledge about it. There is an example repository that explains the model and installation steps. The repository is available at Midas Cpp Example.

The detection uses a simple algorithm, which extracts a particular layer from the calculated estimation and stores it as a binary image. The robot then calculates whether to move left or right by calculating and analyzing the number of pixels on the image.

The aim of the project is to achieve a simple yet fast solution suitable to run on Raspberry Pi 4+ devices using C++ only. This project is developed as my bachelor's degree graduation project "Making an Autonomous Car with Depth Estimation".

Clone this wiki locally