Using Single Shot Multibox Detector (SSD) to detect person and upper of body and then classify each upper clothes with Convolution Neural Network (CNN) to predict multiple label with type and color of upper clothes. There are a total of 10 classes with 3 shirt classes (shirts, t-shirts, jackets) and 7 color classes (black, white, red, gray, yellow, green, blue) equivalent to 2 labels are type and color.
.
├── loss # the localization loss for bounding box offset prediction and the classification
├── models
├── upper.h5 # model detect upper body
├── person.h5 # model detect person
├── upper clothes.h5 # model classify shirt of upper
├── ShirtFashion.ipynb # Combine models to detect and classify shirts
├── Train_Clothes.ipynb # create model upper clothes with CNN
├── detect person and upper.ipynb # create two models detect person and upper with SSD
├── lablebin.pickle # multiple lables for model "upper clothes"
└── README.md
These instructions will get you a copy of the project up and running on your local machine for development and testing purposes. See deployment for notes on how to deploy the project on a live system.
What things you need to install the software and how to install them
Python 3.x
Keras
Tensorflow
Single Shot Multibox Detector With Keras - by author pierluigiferrari
A step by step series of examples that tell you how to get a development env running
Install from PyPI
*Steps:
1. Clone ssd_keras to your computer: !git clone https://github.com/pierluigiferrari/ssd_keras
2. Place files of this project to folder ssd_keras after cloned
3. Run file ShirtFashion.ipynb to review summary of models
This project is licensed under the MIT License - see the LICENSE.md file for details
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent