https://github.com/pyk/hsr

Hand signals recognition using Convolutional Neural Network implemented in TensorFlow
https://github.com/pyk/hsr

Last synced: about 2 months ago
JSON representation

Hand signals recognition using Convolutional Neural Network implemented in TensorFlow

• Host: GitHub
• URL: https://github.com/pyk/hsr
• Owner: pyk
• Created: 2017-01-07T07:20:58.000Z (almost 8 years ago)
• Default Branch: master
• Last Pushed: 2020-04-20T14:03:23.000Z (over 4 years ago)
• Last Synced: 2024-08-02T20:46:48.516Z (4 months ago)
• Language: Python
• Size: 131 KB
• Stars: 108
• Watchers: 9
• Forks: 36
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

• awesome-indo-projects - HSR - Hand signals recognition using Convolutional Neural Network implemented in TensorFlow. (Python)
• awesome-indonesia-repo - HSR - Hand signals recognition using Convolutional Neural Network implemented in TensorFlow. (Python)
• awesome-vietnam-repo - HSR - Hand signals recognition using Convolutional Neural Network implemented in TensorFlow. (Python)

README

        
# HSR

This repository contains code that I use to build a machine learning model for hand signals 

recognition system.

![Eval example](hsr-eval.png)

The training data is not included. You can create your own training data 

using webcam via Chrome. I use the following 

[HTML & JS script](https://gist.github.com/pyk/48b92225d1e3c5a732d1fda7c7b79ce5) 

to collect the training data.

## Running

    python train.py training-data/

It expect all images inside `training-data` directory are named using this 

format: `label_id-*` where `label_id` is natural number and `0 < label_id`.