Creating a model that replicates the functionality of the human eye is challenging. Tasks such as object detection, face recognition, depth calculation, and object tracking are complex and typically require separate machine learning models. In this project, I aim to integrate multiple computer vision models into a single function to emulate the capabilities of the human eye.
Installation and Configuration
The GitHub repository can also be found here:
To install the current release:
python3 -m venv venv
source venv/bin/activate
pip3 install requirements.txt
pip3 freeze > requirements.txt # after add some feature freeze all dependencies
Or:
poetry install
Usage
Move toward the project folder:
cd eye
Run the single function that performs the computer vision tasks:
python inference.py
Hardware Devices
- Coral TPU for object detection.
- Intel RealSense Depth Camera D435 for calculating image depth.
- Face recognition library for face recognition.
- Object tracking custom algorithm for object tracking.