Skip to content

Latest commit

 

History

History
143 lines (83 loc) · 2.89 KB

README.md

File metadata and controls

143 lines (83 loc) · 2.89 KB

Online-Edu-Helper Demo

Table of contents

General info

Sookmyung Women's University 2021 Computer Vision final term project

Team Member : 서희(0hee0), 이도연(omocomo)

Duration of development : 2021.12.15-2021.12.19

Real-time service based on object detection, Online Edu Helper demo

: Detecting face, phone, school supplies in webcam to help remote class management

image

Technologies

Tensorflow Flask React

Custom dataset

  • Goal — To detect face, phone, school supplies in webcam
  • Application — To help remote class management
  • Details — 2200 images with bounding boxes over 9 classes
    • face (400), smartphone (200)
    • ruler, glue, ocarina, recorder, scissors, pen, book (each 200)
    • mix (200)

Install

activate conda

🍎 for m1 mac

install xcode command line tools

xcode-select --install

install Miniforge

https://developer.apple.com/metal/tensorflow-plugin/

chmod +x ~/Downloads/Miniforge3-MacOSX-arm64.sh
sh ~/Downloads/Miniforge3-MacOSX-arm64.sh
source ~/miniforge3/bin/activate

create conda env

  • python version important; python 3.9 or higher is not supported (2021.12)

  • If Anaconda3 is installed, make sure to delete it and check if conda is running on miniforge3 through conda info

conda create —name tf27 python=3.8 
conda activate tf27

install packages

🍎 for m1 mac

install tensorflow

conda install -c apple tensorflow-deps
pip install tensorflow-macos
pip install tensorflow-metal

install object_detection

  1. download the models

    !git clone --q https://github.com/tensorflow/models.git

  1. compiles the proto buffers

    cd models/research

    !protoc object_detection/protos/*.proto --python_out=.

  2. setting up environment variables (Enter a local path to models/ in YOUR_PATH)

    $ vi ~/.bash_profile
    export PYTHONPATH="YOUR_PATH/models:YOUR_PATH/models/slim"
    
    vi ~/.zshrc
    source ~/.bash_profile
    
  3. REBOOT

  4. final install and setup of object detection API

    python setup.py build
    python setup.py install
    

(ModuleNotFoundError: No module named 'tf_slim' pip install --upgrade tf-slim)

install packages for app.py

pip install -r requirements.txt

Usage

Flask

python run.app

React

cd client
yarn
yarn start