license-detection-recognition

Prerequisites

• Python 3
• CPU or NVIDIA GPU + CUDA CuDNN

Getting Started

Installation

As web-service

• Clone this repo

• Create env conda create --name licence-detrec

• Install requirements pip install -r requirements.txt

• Download pretrained model via GD link

• Rename it to medium.pt and put it into detection/models

• Start server: python app.py

As docker (*не получилось пробросить порт и затестить, но сервер запускался :( )

• Run docker build -t <name> path/to/app/dir
• Run docker run -p your_port:5000

Usage

• Send POST request using form or manually

Results

1st row: Input / 2nd row: Result

This service is based on yolov5 link and PaddleOCR link