English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Non-blind deblurring methods achieve decent performance under the accurateblur kernel assumption. Since the kernel uncertainty (i.e. kernel error) isinevitable in practice, semi-blind deblurring is suggested to handle it byintroducing the prior of the kernel (or induced) error. However, how to designa suitable prior for the kernel (or induced) error remains challenging.Hand-crafted prior, incorporating domain knowledge, generally performs well butmay lead to poor performance when kernel (or induced) error is complex.Data-driven prior, which excessively depends on the diversity and abundance oftraining data, is vulnerable to out-of-distribution blurs and images. Toaddress this challenge, we suggest a dataset-free deep residual prior for thekernel induced error (termed as residual) expressed by a customized untraineddeep neural network, which allows us to flexibly adapt to different blurs andimages in real scenarios. By organically integrating the respective strengthsof deep priors and hand-crafted priors, we propose an unsupervised semi-blinddeblurring model which recovers the latent image from the blurry image andinaccurate blur kernel. To tackle the formulated model, an efficientalternating minimization algorithm is developed. Extensive experimentsdemonstrate the favorable performance of the proposed method as compared todata-driven and model-driven methods in terms of image quality and therobustness to the kernel error.

Uncertainty-Aware Unsupervised Image Deblurring with Deep Residual Prior