RoyiAvital / KaggleFacialKeypointsDetection

The Kaggle Facial Keypoints Detection Competition

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Kaggle Facial Keypoints Detection

Overview

Repository of data and solution (MATLAB Based) for The Kaggle Facial Keypoints Detection Competition.
The solutions are using Regression Methods:

  1. Artificial Neural Network.
  2. Random Forest.
  3. SVM Regression.

With Feature Extraction phase:

  1. The Histogram of Gradients (HoG) Feature.
  2. The Local Binary Patterns Feature.

Repository Structure

TBC

How to Run

TBC

ToDo List

List of tasks which might improve performance.

  1. Extend database (Translation).
  2. Investigate why there are no gains by extending database using reflection of images.
  3. Create Feature Specialist Regressor per Keypoint.

About

The Kaggle Facial Keypoints Detection Competition


Languages

Language:MATLAB 43.5%Language:HTML 39.4%Language:C 16.5%Language:Makefile 0.6%