EXPOSEA

is a standalone application designed for composing high-resolution images of large-scale documents. It requires two types of input:

• A captured image of the entire document, referred to as the reference image.
• Multiple close-up images capturing fine details, referred to as fragments.

The process begins by estimating a coarse homography between each fragment and the reference image. Subsequently, a dense optical flow field is computed to achieve a more precise alignment between the pairs. Finally, a light optimization is performed with respect to the reference image to ensure consistent color across composed image form fragments.

Installation

Supported version of Python >= 3.10 and <= 3.12

Clone the repository

 git clone https://github.com/DCGM/Exposea.git
 cd Exposea

Instalation is supported through pip and as a docker image

For Pip

Install requirements

  pip install -r requirements.txt

For Docker

All files required to set up the Docker environment are located in the docker directory.

Start by modifying paths to input and output directories in docker compose

Afterwards you can build the docker as

  cd docker
  docker compose -f docker-compose.yaml build

Run

Docker

You can now either run the one-shot processing mode, which processes every task in the input directory once, or start continuous mode, which keeps running and monitoring for new tasks.

One-shot

  docker compose up exposea

Continuous service

  docker compose up exposea_service

Python script

To run simple example as python script

  python python register.py  -i ./path/to/task -o ./path/to/output/