https://github.com/novatecconsulting/novatec-avatar-generator

Novatec Avatar Generator idea for Hackathon 12.02.2021
https://github.com/novatecconsulting/novatec-avatar-generator

hackathon

Last synced: 2 months ago
JSON representation

Novatec Avatar Generator idea for Hackathon 12.02.2021

• Host: GitHub
• URL: https://github.com/novatecconsulting/novatec-avatar-generator
• Owner: NovatecConsulting
• Created: 2021-02-12T06:28:23.000Z (almost 4 years ago)
• Default Branch: master
• Last Pushed: 2023-10-28T04:30:55.000Z (about 1 year ago)
• Last Synced: 2024-04-18T10:10:38.257Z (9 months ago)
• Topics: hackathon
• Language: JavaScript
• Homepage:
• Size: 1.87 MB
• Stars: 3
• Watchers: 12
• Forks: 2
• Open Issues: 4
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
This application was developed on 12.02.2021 as a submission for Hackathon Event. Contributors: Peter Kutschera, Minh Chi Nguyen.

# git submodules

This repository uses [git submodules](https://git-scm.com/book/en/v2/Git-Tools-Submodules). Please clone this repo with the `--recursive` flag to clone all submodules aswell. An alternative after cloning this repo is `git submodule init && git submodule update`.

# Novatec Avatar Generator Frontend

Angular (11.2) Frontend version for Novatec Avatar Generator offers simple file upload with click or drag and drop. It uses Material Angular and Tailwindcss for styling.

## Installation

`npm install && npm start` will serve at localhost:4200

# Novatec Avatar Generator Backend

The backend is based of U^2 Net, with a Flask server and Image Transformation built on top. The U^2 Net repo can be found at [U^2-Net: Going Deeper with Nested U-Structure for Salient Object Detection](https://github.com/NathanUA/U-2-Net) (Qin et al, Pattern Recognition 2020)

## Installation

- Download the pretrained model [u2net.pth](https://drive.google.com/file/d/1ao1ovG1Qtx4b7EoskHXmi2E9rp5CHLcZ/view)

- Put the file inside the `backend/U-2-Net/saved_models/u2net/` folder, create u2net folder if needed

- Go back to backend folder, install dependencies with pip/pip3  (pip3 for linux/mac)

- Install `pip install -r requirements.txt`. 

- Start the server with `python main.py`