Noise2Noise denoising protocol
The reference implementation that popularised training a denoiser without clean targets. On CT it became the standard 'no clean ground truth' evaluation protocol — most unsupervised CT denoising papers report on a Mayo-LDCT-style split under this setting.
| Type | Benchmark |
|---|---|
| Tasks | denoising, self-supervised |
| Year | 2018 |
| Official page | https://github.com/NVlabs/noise2noise |
Papers in the directory that use or describe it
Papers whose title, note or abstract summary in the directory names this entry, or that link to it.
| Paper | Venue | Year | Links |
|---|---|---|---|
| Noise2Match: Self-Supervised Low-Dose CT Denoising via Adaptive Inter-Slice Matching | preprint | 2026 | paper |
Related topics
Topic pages that list this entry among their datasets and benchmarks.
Guides
Guides that compare this entry with the alternatives.
Everything on this page comes from the directory record (last content update 2026-10-16). Licence terms and access conditions can change: check the official page before you download or redistribute the data.