Prometheus-Vision

Introduction

Prometheus-Vision is a project that utilizes depth cameras to capture depth values from the environment. And to classify objects using vision techniques.

Important

• DO NOT CHANGE any file or file points relating to pybackend,pyrealsense, and pyrsutils. It is related to pyrealsense2 setup.

Usage

Running getDepthValues.py

To run the getDepthValues.py script, follow these steps:

1. Grant Permission: Since the script interacts with the depth camera, you may need to run it with elevated privileges. Use the sudo command as follows: sudo python3 getDepthValues.py

2. Exit the Program: Press the 'q' key while the program is running to exit gracefully.

Dependencies

• Python 3.7
• pyrealsense2 (built from source, and environment is already set up)
• numpy
• tensorflow
• scipy

Dataset used

• NYU Depth Dataset V2
• You may see PotentialDataSetToUse.md for training on more datasets