AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Unsupervised Image Denoising with Score Function

Yutong Xie, Mingze Yuan, Bin Dong, Quanzheng Li

NeurIPS 2023 · Image generation › Fast diffusion sampling · approach: Denoising diffusion probabilistic models · 2 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
Use score-function method instead of Neighbor2Neighbor with correlation transform AGaussian sigma=25 w/ A · KodakPSNR (dB): 27.89 -> 33.99Improved
Use score-function method instead of Noisier2NoiseGaussian sigma=25 w/o A · KodakPSNR (dB): 31.80 -> 31.92No clear effect
Use score-function method instead of Neighbor2Neighbor on signal-dependent Gaussian noise w/o AGaussian a=0.98 b=25 · KodakPSNR (dB): 30.42 -> 29.68Hurt
Use score-function method instead of Neighbor2Neighbor on signal-dependent Gaussian noise w/ AGaussian a=0.98 b=25 · KodakPSNR (dB): 25.02 -> 31.88Improved
Use score-function method instead of Neighbor2Neighbor on unbiased Gamma noiseGamma a=26 w/o A · KodakPSNR (dB): 32.98 -> 32.61Hurt
Use score-function method instead of Neighbor2Neighbor on unbiased Poisson noisePoisson l=0.2 w/o A · KodakPSNR (dB): 32.50 -> 32.38Hurt
Use score-function method instead of Neighbor2Neighbor on biased Rayleigh noiseRayleigh s=0.3 w/o A · KodakPSNR (dB): 16.55 -> 34.25Improved
Use score-function method instead of Neighbor2Neighbor on mixture Gamma noise w/ AGamma a=26 + Gaussian, w/ A · KodakPSNR (dB): 26.93 -> 30.74Improved
Use score-function method instead of Neighbor2Neighbor on mixture Poisson noise w/o APoisson l=0.2 + Gaussian, w/o A · KodakPSNR (dB): 32.03 -> 32.10No clear effect
Use score-function method instead of Neighbor2Neighbor on mixture Gamma noise w/o AGamma a=26 + Gaussian, w/o A · KodakPSNR (dB): 32.30 -> 32.13Hurt
Use score-function method instead of Neighbor2Neighbor on mixture Rayleigh noise w/ ARayleigh s=0.3 + Gaussian, w/ A · KodakPSNR (dB): 16.40 -> 31.14Improved
Replace supervised learning with unsupervised score-function method on Gaussian noiseGaussian sigma=25 w/o A · KodakPSNR (dB): 32.44 -> 31.92Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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