Stabilizing Deep Reconstruction Operators with Contractive Anchoring

• URL: http://arxiv.org/abs/2607.23341v2
• Date: Wed, 29 Jul 2026 21:04:29 GMT
• Title: Stabilizing Deep Reconstruction Operators with Contractive Anchoring
• Abstract summary: Image Reconstruction (IR) regularizers in an iterative process can destabilize reconstruction.<n>A common failure mode is the peak-and-collapse: metrics such as PSNR improve for early behaviour and then abruptly degrade, making these algorithms unreliable in practice.<n>We propose a data-driven stabilization framework that formalizes this instability through a local quantity.<n>Our key idea is to control the potentially unstable IR operator with a contractive operator whose stable iterations act as an anchor and prevent collapse.
• Score: 12.504926215719335
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Pretrained deep denoisers can be used to solve a wide range of model-based image reconstruction tasks via Plug-and-Play (PnP) and Regularization-by-Denoising (RED) algorithms, without retraining per task. These denoisers are trained only for single-step denoising. Using them as Image Reconstruction (IR) regularizers in an iterative process can destabilize reconstruction. A common failure mode is the peak-and-collapse behaviour: metrics such as PSNR improve for early iterations and then abruptly degrade, making these algorithms unreliable in practice. We propose a data-driven stabilization framework that (i) formalizes this instability of any IR operator through a local quantity and (ii) prevents collapse by regularizing this quantity adaptively, requiring no retraining or modification of the given pretrained network. Our key idea is to control the potentially unstable IR operator with a contractive operator whose stable iterates act as an anchor and prevent collapse. We further introduce an efficient family of trainable contractive operators that serve as strong anchors while remaining lightweight. Extensive experiments across proximal algorithms, denoiser architectures, noise levels, and imaging tasks show consistent, collapse-free performance and improved reliability of PnP and RED reconstruction.

Related papers

• Frozen CLIP Priors for Robust Self-Supervised Poisson Inverse Problems [19.52206071875139]
  This paper proposes an ADMM-inspired unrolled plug-and-play solver for Poisson inverse problems that decouples a closed-form data-consistency update from a parameter-efficient prior.<n> Experiments on Poisson CFA demosaicing and deblurring show competitive quality, improved robustness under shifts, and self-supervised performance approaching supervised training.
  arXiv  Detail & Related papers  (2026-08-20T19:37:52Z)
• StructRL: Structured Action-Space Exploration for Flow-Based VLAs [81.43458629514139]
  Flow-based Vision-Language-Action models are now widely used for continuous robotic manipulation.<n>Online reinforcement learning is emerging as a key technique for adapting them to new tasks.<n>We show that effective robot exploration calls for structured noise: temporally smooth and scaled differently across action groups.
  arXiv  Detail & Related papers  (2026-08-15T09:31:29Z)
• Noise-Free One-Step LoRA for Task-Driven Image Restoration with Diffusion Priors [54.200377321709546]
  Degraded images not only reduce visual quality but also impair downstream high-level vision tasks.<n> Task-driven image restoration (TDIR) addresses this issue by jointly optimizing restoration quality and task performance.<n>We show that a deterministic, noise-free one-step forward pass with pretrained diffusion priors can substantially improve TDIR.<n>We also introduce a task-preserving GAN training strategy that improves perceptual quality without sacrificing task performance.
  arXiv  Detail & Related papers  (2026-07-28T07:51:44Z)
• SATB-VR: Training Few-Step Video Restoration Diffusion Model using SNR-Aware Trajectory Blending [55.764018145256216]
  We present SATB-VR, a few-step paradigm that jump-starts the denoising process via an auxiliary predictor.<n>We show that SATB-VR performs favorably against existing approaches on synthetic, real-world, and AIGC benchmarks.
  arXiv  Detail & Related papers  (2026-06-27T01:32:16Z)
• Clipping Bottleneck: Stabilizing RLVR via Stochastic Recovery of Near-Boundary Signals [83.0127582612634]
  Near-boundary Rescue (NSR) is a minimal, plug-and-play modification that retains out-of-bound tokens to recover lost signals.<n>NSR substantially improves training stability and delivers consistent gains over strong baselines such as DAPO and GSPO.
  arXiv  Detail & Related papers  (2026-05-21T16:45:31Z)
• Next-Acceleration-Scale Prediction for Autoregressive MRI Reconstruction [52.32112533846212]
  MRI reconstruction is an inherently ill-posed inverse problem, since incomplete measurements admit many plausible solutions.<n>We address this limitation by moving reconstruction to discrete multi-scale latent space and posing it as autoregressive next-acceleration-scale prediction.<n>We show that our approach delivers improved reconstruction performance across diverse sampling patterns under extreme undersampling.
  arXiv  Detail & Related papers  (2026-05-19T04:40:50Z)
• Provably Contractive and High-Quality Denoisers for Convergent Restoration [4.912983462300839]
  Existing convolutional and attention-based networks lack stability guarantees under minor shifts in input.<n>We develop provably contractive (global Lipschitz $ 1$) denoiser networks that considerably reduce this gap.<n>Our results show that enforcing strict Lipschitz control does not inherently degrade output quality.
  arXiv  Detail & Related papers  (2026-03-27T08:35:55Z)
• Robust Stochastically-Descending Unrolled Networks [85.6993263983062]
  Deep unrolling is an emerging learning-to-optimize method that unrolls a truncated iterative algorithm in the layers of a trainable neural network.<n>We show that convergence guarantees and generalizability of the unrolled networks are still open theoretical problems.<n>We numerically assess unrolled architectures trained under the proposed constraints in two different applications.
  arXiv  Detail & Related papers  (2023-12-25T18:51:23Z)
• Advancing Unsupervised Low-light Image Enhancement: Noise Estimation, Illumination Interpolation, and Self-Regulation [55.07472635587852]
  Low-Light Image Enhancement (LLIE) techniques have made notable advancements in preserving image details and enhancing contrast. These approaches encounter persistent challenges in efficiently mitigating dynamic noise and accommodating diverse low-light scenarios. We first propose a method for estimating the noise level in low light images in a quick and accurate way. We then devise a Learnable Illumination Interpolator (LII) to satisfy general constraints between illumination and input.
  arXiv  Detail & Related papers  (2023-05-17T13:56:48Z)
• DiffusionAD: Norm-guided One-step Denoising Diffusion for Anomaly Detection [80.20339155618612]
  DiffusionAD is a novel anomaly detection pipeline comprising a reconstruction sub-network and a segmentation sub-network.<n>A rapid one-step denoising paradigm achieves hundreds of times acceleration while preserving comparable reconstruction quality.<n>Considering the diversity in the manifestation of anomalies, we propose a norm-guided paradigm to integrate the benefits of multiple noise scales.
  arXiv  Detail & Related papers  (2023-03-15T16:14:06Z)
• Learning Sparsity-Promoting Regularizers using Bilevel Optimization [9.18465987536469]
  We present a method for supervised learning of sparsity-promoting regularizers for denoising signals and images. Experiments with structured 1D signals and natural images show that the proposed method can learn an operator that outperforms well-known regularizers.
  arXiv  Detail & Related papers  (2022-07-18T20:50:02Z)
• Supervised Learning of Sparsity-Promoting Regularizers for Denoising [13.203765985718205]
  We present a method for supervised learning of sparsity-promoting regularizers for image denoising. Our experiments show that the proposed method can learn an operator that outperforms well-known regularizers.
  arXiv  Detail & Related papers  (2020-06-09T21:38:05Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.