AI Paper Atlas

Research map / Label-efficient and robust learning / Noisy labels

Evaluating Unsupervised Denoising Requires Unsupervised Metrics

Adria Marcos Morales, Matan Leibovich, Sreyas Mohan, Joshua Lawrence Vincent, Piyush Haluai, Mai Tan, Peter Crozier, Carlos Fernandez-Granda

ICML 2023 · Label-efficient and robust learning › Noisy labels · approach: Contrastive learning on noisy data · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Estimate PSNR with uPSNR using independent noisy references from the same clean imageDenoisers on natural images with Gaussian noise · Natural images, Gaussian noise σ=25, 50, 75, and 100PSNR agreement: 26.19 -> 26.21; 23.95 -> 24.02; 22.72 -> 22.75; 21.84 -> 21.84No clear effect
Use spatially subsampled references from a single noisy image instead of references with the same underlying clean image (ablation)uPSNR · Natural images, Gaussian noisePSNR agreement: 24.20 -> 26.20; 26.54 -> 27.61; 26.19 -> 28.14; 27.22 -> 25.40Hurt
Use spatially subsampled references from a single noisy image (ablation)uPSNR · Electron microscopy, Poisson noisePSNR agreement: 20.18 -> 20.21; 24.86 -> 24.74; 25.74 -> 25.86; 24.65 -> 24.79No clear effect
Estimate PSNR using uPSNR from three noisy references rather than averaging-based PSNR from ten noisy referencesDenoisers on RAW videos with real noise · RAW videos, all ISO levelsPSNR estimate: 32.18 -> 32.27; 42.99 -> 43.03; 30.47 -> 30.49; 44.72 -> 44.80No clear effect
Use spatial subsampling when the image content is not sufficiently smooth at the pixel resolution (ablation)uMSE and uPSNR · Single-image evaluation of natural imagesmetric biasHurt
Increase image smoothness before deriving references through spatial subsampling (ablation)uMSE and uPSNR · Controlled natural-image experimentuMSE errorImproved
Use three independent noisy references and subtract the pairwise-reference noise correctionuMSE · Signals satisfying independence and centered-noise conditionsMSE estimationImproved
Use uPSNR derived from uMSE rather than requiring clean ground truthDenoising evaluation · Noisy-data evaluation under the theorem assumptionsPSNR estimationImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy