Skip to content

Latest commit

 

History

History
58 lines (34 loc) · 2.18 KB

README.md

File metadata and controls

58 lines (34 loc) · 2.18 KB
Raw

Optical Flow Computation (Horn-Schunck Method)

Overview

This is a Python implementation of the Horn-Schunck optical flow computation method. Given an image sequence as input, this program calculates flow vectors (u, v) that represent pixel motion between consecutive frames. The Horn-Schunck method is a classic and widely used approach for estimating optical flow in computer vision.

Example usage: Test image and corresponding result image

Test Image Result Image

Features

  • Consider two input images of consecutive frames, and compute optical flow.
  • Utilizes the Horn-Schunck method for optical flow estimation.
  • Regularization constant (alpha) to control the smoothness of the output flow vectors.
  • Easily adjustable parameters for customized results.

Installation

Before running the program, make sure you have Python (>= 3.6) installed on your system. Clone this repository and install the required dependencies:

git clone https://github.com/chanukaravishan/Optical-flow-Horn-Schunk.git
cd Optical-flow-Horn-Schunk
pip install -r requirements.txt

Usage

To compute optical flow for your image sequences, follow these steps:

  1. Prepare your image sequences: Make sure you have consecutive image frames in a directory.

  2. Open the of_hornschunk.py script and set the alpha parameter according to your requirements. The higher the value of alpha, the smoother the output flow vectors will be.

  3. Run the script by providing the path to your image sequence directory:

python of_hornschunk.py --input_path /path/to/your/image_sequences --output_path /path/to/save/flow_results

Contributing

Contributions to this project are welcome! Feel free to open issues and submit pull requests to address bugs, add new features, or improve the existing implementation.

Credits

This optical flow computation program was developed by Chanuka Algama (https://github.com/ChanukaRavishan).

Acknowledgments

Special thanks to the original authors of the Horn-Schunck method for their valuable research and contribution to the computer vision community.