/Hand-Gesture-Recognition

A computer vision and machine learning pipeline that classifies images of Colombian hand gestures (0 to 5).

Primary LanguageJupyter Notebook

Hand Gesture Recognition

Run on test images

  1. Put your test images in data.
  2. Run 
    python src/main.py ../data ../results
  3. See the results in results

It is better to rename images as 1, 2, 3 and so on.

Requirements

pip install -r requirements.txt

Directory Structure

Hand-Gesture-Recognition/
├───models
├───features
├───dataset
├───results
│   ├───results.txt
│   ├───time.txt
├───src
│   ├───main.py
│   ├───feature_extraction
│   ├───model_training
│   ├───pre-process
│   └───utils
│      └───common_functions.py
│      └───constants.py
├───.gitignore
├───project_requirements.pdf
├───project_report.pdf
├───README.md
├───requirements.txt

Dataset

This drive directory contains:

  • Dataset_0-5.zip: containes Men and Women directories in which you find a directory of images for each number from 0 to 5.
  • dataset_sample.zip: sample images from the original dataset.

Download and move Dataset_0-5.zip after unzipping it into a directory dataset. If it does not exist, create it.

Members