PyFlask_DL-service

Flask for Computer-Vision Prototype

Link : http://54.180.150.154:5000/ (현재 서버는 내렸습니다!😅)

Function

• Object Detection by Darkflow with Fast YOLO-V2
• Neural Style Transfer

Setting Virtualenv

1. Create virtualenv in Anaconda

> conda create -n venv_flask python=3.7

2. Activate virtualenv

> conda activate venv_flask

Install Requirements

1. Install flask & library

(venv_flask) > cd ~pyflask
(venv_flask) ~pyflask > pip install -r requirements.txt

• Flask==1.1.1
• pillow==7.0.0
• opencv-python==4.1.2.30
• requests==2.22.0
• tensorflow-gpu==1.15.0
• Keras==2.3.1
• Cython==0.29.14
• darkflow==1.0.0

2. Install Darkflow

(venv_flask) ~pyflask > cd .\flask_deep\darkflow_yolo
(venv_flask) ~pyflask\flask_deep\darkflow_yolo > pip install .

Start Server

(venv_flask) ~pyflask\flask_deep\darkflow_yolo > cd ../
(venv_flask) ~pyflask\flask_deep > cd ../
(venv_flask) ~pyflask > python start_flask.py

Main Page

ConvNet? Page

Object Detection Page

• Input

• Output

Neural Style Transfer

• Input

• Output