Skip to content
PL group project desktop app - done by Urmil Shroff, Vinay Kolwankar and Priyansh Ramnani👟
Python HTML JavaScript Shell CSS
Branch: master
Clone or download
Latest commit 63802f2 Mar 27, 2019
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
backend Removed useless 3rd page and some minor improvements Mar 21, 2019
css Display selected image Feb 23, 2019
datasets Added two more shoes Mar 21, 2019
demo Added screenshots and demo videos Mar 10, 2019
img Optimized splash screen Mar 20, 2019
inception Revert "Retrained entire model" Mar 5, 2019
test_images Final Mar 27, 2019
tf_files Final Mar 27, 2019
.gitignore Merge branch 'nike' into urmil-gui Mar 12, 2019
LICENSE Initial commit Jan 12, 2019
README.md Update README.md Mar 10, 2019
electron_packager.sh Prepare for packaging Mar 5, 2019
image_select.html Removed useless 3rd page and some minor improvements Mar 21, 2019
index.html Moved solefinder.js to /backend Mar 10, 2019
install.sh Improved installation Feb 15, 2019
loading_screen.html Removed useless 3rd page and some minor improvements Mar 21, 2019
main.js Optimized splash screen Mar 20, 2019
package-lock.json
package.json
requirements.txt Added script for basic webscraping Mar 10, 2019
retrain.sh Updated Mar 12, 2019
run.sh Final Mar 27, 2019

README.md

SoleFinder

SoleFinder

A Deep Learning app made using TensorFlow and Electron.js to detect the exact model of different pairs of shoes and provide information about them.

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.

Prerequisites

  • Python 3.6.8
  • Node.js 6.9.0

Installing

Building

  • Clone this repo using git clone https://github.com/urmilshroff/solefinder.git
  • Type cd solefinder to go into the repo directory
  • Run bash install.sh to install all the dependencies
  • Run npm start to launch the app

Retraining Dataset

If you wish to add your own shoes/create your own dataset, then use the following method.

  • Download 100-150 images of a particular shoe
  • Save them all in /datasets/shoe_name/
  • Run python /datasets/shoe_rename.py to rename all the images
  • Run bash retrain.sh to retrain the dataset

Contributing

This project is licensed under the MIT license. Please feel free to clone/fork/modify the code as you like, and send us pull requests for any improvements, bugfixes or features. We'll definitely look into them and merge them if satisfactory.

Authors

Also checkout the list of contributors who participated in this project.

License

This project is licensed under the MIT License - see the LICENSE file for details.

You can’t perform that action at this time.