/face_detection

Python Jupyter Notebook Application that detects faces in the pictures using HOG + SVM.

Primary LanguageJupyter Notebook

FACE DETECTION

Algorithm: HOG (Histogram of Oriented Gradients) + SVM (Support Vector Machine)

Dataset: 7 500 pictures of Faces and 7 500 pictures of different objects

Main libraries: numpy and opencv

Includes own implementation of HOG and Canny

INSTRUCTION

To run this program simply download anaconda python distribution, activate anaconda virtual env, run jupyter notebook and click run all option.

Important:

You must have installed following packages: opencv-python, matplotlib, numpy, pandas, tqdm.

SAMPLE RESULTS

alt text

alt text

alt text

alt text

alt text

alt text