Jul 2026· International Conference on Information Photonics· Vol abs/2607.21078, pp. 1-6· 0 citations· 32 references
Computer Science
TL;DR
RealDefocus, a benchmark derived from the real-world RealBokeh dataset originally proposed for Bokeh Rendering, is built on, providing paired defocused inputs and sharp ground truth images, predefined training/validation/test splits, and a unified evaluation framework for comparing image restoration and neural rendering approaches.
Abstract
Single-Image Defocus Deblurring (SIDD) aims to recover an all-in-focus image from a single defocused observation, but rigorous and reproducible evaluation remains challenging due to the scarcity of realistic, high-resolution datasets with well-aligned defocused/sharp pairs and standardized protocols. We build on RealDefocus, a benchmark derived from the real-world RealBokeh dataset originally proposed for Bokeh Rendering. RealDefocus provides paired defocused inputs and sharp ground truth images, predefined training/validation/test splits, and a unified evaluation framework for comparing image restoration and neural rendering approaches. We further outline a benchmarking protocol with cross-dataset validation to assess reconstruction quality and generalization. The project page is publicly available at: www.github.com/TimSeizinger/RealDefocus-Benchmark.
Defocus deblurring is a challenging task due to spatially varying blur and limited aligned training data. Existing datasets suffer from insufficient scene diversity and misalignment between defocused and all-in-focus images, restricting network performance. Additionally, single-stage autoencoders often fall into local...
Yiming Zhong, Ji-Feng Chen· International Conference on...· 0 citations
The general ideas, typical examples and applications of the four main architectural styles, as well as their particular advantages and deficiencies in different blurring conditions, are introduced.
Yuxuan Wu· Applied and Computational En...· 0 citations
Image acquisition with a camera involves several degradations due to the optical system, sensor, or low-level processing steps. We address blind deblurring in professional photography: we aim to invert unknown isotropic blur without knowledge of the degradation kernel. For such inverse problems,where some high-frequenc...
Imane Si Salah, Emile Cribelier, T. Veit et al.· International Conference on...· 0 citations
This paper proposes a novel outpainting technique for quadri-wave lateral shearing interferograms which employs Quadri-Wave Generative Adversarial Networks (QW-GANs). QW-GAN is adversarially trained to capture the high-order statistics of real interferograms. It thereby avoids the Gibbs-ringing artifacts and spectral l...
Yao Fan, Ya-Xuan Duan, Yi-Wen Zhang et al.· Italian National Conference...· 0 citations
PI-NAFDeblur is presented, a physics-informed neural network that utilizes a coupling of a lightweight NAFNet restoration backbone and a differentiable reblur operator that enforces forward-model consistency during training to demonstrate that physics-informed regularization can help improve cross-dataset robustness un...
Riam H. Saffar· Al-Noor Journal of Engineeri...· 0 citations
Field-dependent blur and frequency-response loss limit miniature confocal endoscopic imaging. We introduce a physics-prior-guided dual-branch residual network that combines the degraded image with normalized field coordinates and calibration-derived MTF, PSF, SNR, and intensity-response maps. For the representative per...
Zhi Wang, Xue-Yi Wang, Yi-Ning Mu et al.· Optics Express· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
