Search arXivSearch

arXiv · 2510.16428

Dictionary-Based Deblurring for Unpaired Data

Abstract

Effective image deblurring typically relies on large and fully paired datasets of blurred and corresponding sharp images. However, obtaining such accurately aligned data in the real world poses a number of difficulties, limiting the effectiveness and generalizability of existing deblurring methods. To address this scarcity of data dependency, we present a novel dictionary learning based deblurring approach for jointly estimating a structured blur matrix and a high resolution image dictionary. This framework enables robust image deblurring across different degrees of data supervision. Our method is thoroughly evaluated across three distinct experimental settings: (i) full supervision involving paired data with explicit correspondence, (ii) partial supervision employing unpaired data with implicit relationships, and (iii) unsupervised learning using non-correspondence data where direct pairings are absent. Extensive experimental validation, performed on synthetically blurred subsets of the CMU-Cornell iCoseg dataset and the real-world FocusPath dataset, consistently shows that the proposed framework has superior performance compared to conventional coupled dictionary learning approaches. The results validate that our approach provides an efficient and robust solution for image deblurring in data-constrained scenarios by enabling accurate blur modeling and adaptive dictionary representation with a notably smaller number of training samples.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Alok Panigrahi, Jayaprakash Katual, Satish Mulleti. 2025-10-18. Dictionary-Based Deblurring for Unpaired Data. https://arxiv.org/abs/2510.16428

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

KELP: K-space-conditioned Estimation of Learned Sampling Patterns for Scan-Adaptive Multi-Coil MRI

Deep learning techniques have gained considerable attention for accelerating MRI acquisition while maintaining image quality. In this work, we present a convolutional neural network (CNN)-based framework for predicting scan-adaptive undersampling patterns directly from low-frequency multi-coil $k$-space data. Unlike approaches that optimize sampling patterns during training or rely on nearest-neighbor search at inference, our method is trained using precomputed scan-adaptive optimized masks as supervised labels and predicts a scan-specific sampling pattern in a single forward pass. The training procedure alternates between optimizing the sampling network and a reconstruction network, allowing the learned masks to adapt to reconstruction performance. Experiments on the fastMRI multi-coil knee dataset demonstrate competitive reconstruction quality compared with existing population- and scan-adaptive sampling approaches at $4\times$ and $8\times$ acceleration factors. In addition, the proposed method enables efficient scan-specific mask prediction with inference cost independent of the training dictionary size.

eess.IV

LaminoDiff: Generative Computed Laminography via Near-Isotropic Spectral Supervision and Anisotropic Geometry

Computed Laminography (CL) is widely used for nondestructive inspection of extended planar objects, but its restricted angular coverage produces an anisotropic point spread function, missing-cone spectral incompleteness, and severe aliasing with interlayer leakage. This paper presents LaminoDiff, a physics-constrained diffusion framework for CL reconstruction. During training, a CT-derived near-isotropic supervision target is reconstructed from full-angle Computed Tomography (CT) projections physically degraded to match CL detector noise and focal-spot blur while retaining full angular coverage; it is withheld from inference. At inference, the reverse process uses only the CL observation. An Anisotropic Representation (AR) constructs depth-aware channels from three adjacent slices: a neighbor average, a center-neighbor residual, and the retained current slice, followed by directional in-plane feature extraction. Experiments on simulated and real multilayer printed circuit board data, including ball grid array and high-frequency stub samples, show that LaminoDiff improves artifact suppression, edge preservation, and depth stratification over analytic Feldkamp--Davis--Kress and representative learning-based baselines.

eess.IV

IViT: A Novel Interpretable Visual Transformer for Skin Disease Detection

The clinical diagnosis of skin diseases is susceptible to interference from inter-class similarity of skin lesions, and over-reliance on clinicians'experience easily leads to subjective bias. Although existing deep learning aided diagnosis methods achieve competitive accuracy, they suffer from the black-box opacity of Vision Transformer (ViT) and poor adaptability to medical few-shot scenarios. Moreover, mainstream explainable algorithms generally face the bottleneck of significant accuracy degradation when improving interpretability. This paper proposes an interpretable ViT (IViT) constrained by Quadratic Programming (QP). The introduced pre-trained transfer learning adapts to few-shot feature extraction. A discrete QP feature selection framework is constructed to screen generic and discriminative features consistent with clinical diagnostic logic. A multi-objective loss function is designed to reduce feature redundancy and optimize activation distribution while preserving classification performance. Experimental results on six standard skin disease datasets show that IViT achieves an accuracy of 93.80%, only 0.21% lower than the baseline, with feature redundancy reduced by 29.5%. Its core activation regions are consistent with clinically concerned lesion areas. The proposed model balances accuracy and interpretability, providing a reliable solution for the clinical deployment of few-shot intelligent skin disease diagnosis.

eess.IV