Skip to content

robcog-iai/semlog_web

Repository files navigation

SemLog Web

SemLog Web is an integrated framework written in Python with a concise web UI that bridges between knownledge-based virtual environment and numerous computer vision models.

Getting Started

Go to "semlog_web/web/example" and click "DEMO.ipynb" to directly get a picture of SemLog Web's main features. You can also install dependencies to deploy the framework locally

Prerequisites

You need to retrieve an instance of SemLog database to use this framework.

Installing

  1. Install Miniconda - Python 3.7 from https://docs.conda.io/en/latest/miniconda.html
  • In "Advanced options", tick both options (Add Anaconda to my PATH environment)
  • Note: You can test if Python is installed by enter "python" in your terminal.
  1. clone the repo
git clone https://github.com/robcog-iai/semlog_web.git
  1. Open a terminal with the project root path, run
pip install -r requirements.txt

To install all required libraries.

Install pytorch to enable real time model training. If you don't have a GPU, simply run

pip install torch==1.4.0+cpu torchvision==0.5.0+cpu -f https://download.pytorch.org/whl/torch_stable.html
  1. make sure the local MongoDB is running.

  2. In the terminal, run

python manage.py runserver

To start the server.

  1. Open your browser and visit the website via the address "localhost:8000".

Deployment

Open your browser and visit the website via the address "localhost:8000" to deploy it locally, or use

python manage.py runserver TARGET_IP

To deploy the website in local network

Built With

  • Django - The web framework used
  • MongoDB - The database used
  • Pytorch - The deep learning framework used to build the detection demo
  • Three.js - The UI JavaScript framework used for dynamic point cloud generation.

Authors

  • Xiaoyue Zhang - Author - xinliy

License

This project is licensed under the BSD-3-Clause License

Acknowledgments

  • This framework is developed by Xiaoyue Zhang in the Institute of Artificial Intelligence, University of Bremen, under the supervision of Andrei Haidu.