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

Single image denoising (SID) has achieved significant breakthroughs with thedevelopment of deep learning. However, the proposed methods are oftenaccompanied by plenty of parameters, which greatly limits their applicationscenarios. Different from previous works that blindly increase the depth of thenetwork, we explore the degradation mechanism of the noisy image and propose alightweight Multiple Degradation and Reconstruction Network (MDRN) toprogressively remove noise. Meanwhile, we propose two novel HeterogeneousKnowledge Distillation Strategies (HMDS) to enable MDRN to learn richer andmore accurate features from heterogeneous models, which make it possible toreconstruct higher-quality denoised images under extreme conditions. Extensiveexperiments show that our MDRN achieves favorable performance against other SIDmodels with fewer parameters. Meanwhile, plenty of ablation studies demonstratethat the introduced HMDS can improve the performance of tiny models or themodel under high noise levels, which is extremely useful for relatedapplications.

Multiple Degradation and Reconstruction Network for Single Image Denoising via Knowledge Distillation