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

Due to their highly structured characteristics, faces are easier to recoverthan natural scenes for blind image super-resolution. Therefore, we can extractthe degradation representation of an image from the low-quality and recoveredface pairs. Using the degradation representation, realistic low-quality imagescan then be synthesized to fine-tune the super-resolution model for thereal-world low-quality image. However, such a procedure is time-consuming andlaborious, and the gaps between recovered faces and the ground-truths furtherincrease the optimization uncertainty. To facilitate efficient model adaptationtowards image-specific degradations, we propose a method dubbed MetaF2N, whichleverages the contained Faces to fine-tune model parameters for adapting to thewhole Natural image in a Meta-learning framework. The degradation extractionand low-quality image synthesis steps are thus circumvented in our MetaF2N, andit requires only one fine-tuning step to get decent performance. Consideringthe gaps between the recovered faces and ground-truths, we further deploy aMaskNet for adaptively predicting loss weights at different positions to reducethe impact of low-confidence areas. To evaluate our proposed MetaF2N, we havecollected a real-world low-quality dataset with one or multiple faces in eachimage, and our MetaF2N achieves superior performance on both synthetic andreal-world datasets. Source code, pre-trained models, and collected datasetsare available at https://github.com/yinzhicun/MetaF2N.

MetaF2N: Blind Image Super-Resolution by Learning Efficient Model Adaptation from Faces