Ground Truth Free Denoising by Optimal Transport
This addresses the challenge of denoising in scenarios where ground truth data is unavailable, though it appears incremental as it builds on existing WGAN frameworks.
The authors tackled the problem of unsupervised denoising for arbitrary data types without requiring paired noisy-clean samples, achieving results solely from noisy data and independent additive noise examples.
We present a learned unsupervised denoising method for arbitrary types of data, which we explore on images and one-dimensional signals. The training is solely based on samples of noisy data and examples of noise, which -- critically -- do not need to come in pairs. We only need the assumption that the noise is independent and additive (although we describe how this can be extended). The method rests on a Wasserstein Generative Adversarial Network setting, which utilizes two critics and one generator.