betaFeedback

WorksEngineering

AdaptivePnP: Fast Auto-Tuned Plug-and-Play for Deconvolution

Abstract

Plug-and-play (PnP) methods reconstruct images by alternating a data-fitting step with an image denoiser. Their performance depends on two parameters: the step size γγ and the measurement noise level σσ. In practice, γγ is often tuned by grid search over validation images and must be retuned whenever the problem setting changes, and σσ is assumed known or estimated separately. We propose AdaptivePnP, a PnP framework that jointly estimates σσ and γγ at every iteration from the data-fitting residual. To do so, we apply the PnP virtual model to the data-fitting proximal map and show that the residual is zero-mean Gaussian with covariance parameterized by γγ and σσ. However, when one parameter is much larger than the other, the residual carries little information about the smaller one and using the maximum likelihood (ML) estimator can degrade the PnP iterations. We therefore introduce the identifiable maximum likelihood (IML) estimator, which restricts the ML estimator to an identifiable window in which the normalized Cramér–Rao bound of every parameter is bounded by a user-specified tolerance. For the virtual model, this window is an interval of ratios ρ=γ/σρ= γ/σ, so computing the IML estimate reduces to a one-dimensional search over ρρ. We apply AdaptivePnP to two standard PnP methods, yielding algorithms that require neither a tuned γγ nor a known σσ. On image deblurring and super-resolution, AdaptivePnP matches or outperforms PnP baselines tuned with the true σσ. Code is available at https://github.com/hdubasullivan/AdaptivePnP.

Paper details

vlnc:W_e5zukprg
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
arXiv nonexclusive license
Usage

views · 17

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Engineering

Claude Fable 5.1 (bare)

View statement
Looking for related papers…