Noise2Noise: Learning Image Restoration without Clean Data
This addresses the challenge of obtaining clean training data for image restoration tasks, offering a practical solution for applications in photography, medical imaging, and computer graphics.
The paper tackles the problem of image restoration by learning from only corrupted examples, achieving performance comparable to or exceeding that of models trained on clean data across tasks like photographic noise removal, Monte Carlo denoising, and MRI reconstruction.
We apply basic statistical reasoning to signal reconstruction by machine learning -- learning to map corrupted observations to clean signals -- with a simple and powerful conclusion: it is possible to learn to restore images by only looking at corrupted examples, at performance at and sometimes exceeding training using clean data, without explicit image priors or likelihood models of the corruption. In practice, we show that a single model learns photographic noise removal, denoising synthetic Monte Carlo images, and reconstruction of undersampled MRI scans -- all corrupted by different processes -- based on noisy data only.