Zoe Depth Estimation

This project converts 2D images into detailed depth maps using advanced machine learning and computer vision techniques. The generated depth maps provide per-pixel distance information, which is essential for applications like augmented reality, 3D scene reconstruction, and automated object detection.

Features:

• Implements state-of-the-art algorithms for single-image depth estimation.
• Facilitates the understanding of spatial structures in scenes.
• Serves as a basis for developing enhanced computer vision and AR applications.

Usage/Examples

CLI Usage

usage: CLI.py [-h] input_image output_image

Depth Estimation Model CLI.

positional arguments:
    input_image    Path to the input image.
    output_image   Path to the output depth map.

options:
  -h, -help        show this help message and exit

API Usage

uvicorn API:app

API hosted at

http://127.0.0.1:8000/docs

Installation

Install depth estimation requirements with pip

pip install -r requirements.txt

Environment Variables

To run this project, add the following environment variable to your .env file:

• API_KEY: Visit imgbb API to obtain your API key.

License

MIT