fda 0.0.2

Package for doing a simple Fourier-based domain adaptation.

Fourier Domain Adaption (FDA)

This Python package implements a classic frequency domain adaptation, as shown in:

FDA: Fourier Domain Adaptation for Semantic Segmentation, Yanchao Yang and Stefano Soatto, CVPR 2020

Install with pip

$ python3 -m pip install fda --user

Install from source

$ python3 setup.py install --user

Exemplary code snippet

import fda

# Read source and target images
source_im = cv2.imread('source.jpg')
target_im = cv2.imread('target.jpg')

# Perform domain adaptation
adapted_im = fda.fda(source,_im, target_im, beta=0.005)

Run domain adaptation on a single image

$ python3 -m fda.run --source source.jpg --target target.jpg --output output.jpg --beta 0.005

Some examples of the domain adaptation

Source image	Target domain image	Beta	Output
		0.001
		0.01
		0.1
		0.001
		0.01
		0.1
		0.001
		0.01
		0.1

Run unit tests

$ python3 tests/test_fourier.py

License

This repository is shared under an MIT license.

Author

Luis Carlos Garcia Peraza Herrera ([e-mail]), 2020-2022.