Welcome to the HOG3D Visualization repository! This project provides an interactive web interface to visualize the implementation and results of the Histogram of Oriented Gradients 3D (HOG3D) algorithm, particularly for detecting blocks in coronary arteries using medical imaging data in NIFTI format.
- NumPy: Efficient computation of gradient histograms in 3D.
- Plotly: Interactive 3D plots for visualizing HOG3D features.
- HOG3D.ipynb: Jupyter Notebook demonstrating the implementation of HOG3D.
- HOG3D.pdf: Detailed explanation of the HOG3D implementation.
-
Clone the repository:
git clone https://github.com/Pranav-Karra-3301/HOG3D-Visualization.git cd hog3d -
Install the required Python packages:
pip install numpy plotly nibabel scipy
Open HOG3D.ipynb in Jupyter Notebook to explore the implementation of the HOG3D algorithm.
Explore the detailed implementation of the HOG3D algorithm in the Jupyter Notebook. View Notebook
Interact with the 3D plots generated using Plotly. View Graphs
Check out the source code and contribute to the project. View GitHub
The HOG3D (Histogram of Oriented Gradients 3D) implementation is developed using NumPy to facilitate efficient computation of gradient histograms in 3D. This implementation is designed to work with medical imaging data, particularly for detecting blocks in coronary arteries. The notebook provided demonstrates how to apply the HOG3D algorithm to NIFTI (.nii or .nii.gz) files, which are commonly used in medical imaging.
For a detailed explanation of the HOG3D implementation, refer to the provided PDF document below. It includes step-by-step instructions and insights into the algorithm.
Your browser does not support PDFs. Download the PDF.