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

Although deep learning has significantly improved Face Recognition (FR),dramatic performance deterioration may occur when processing Low Resolution(LR) faces. To alleviate this, approaches based on unified feature space areproposed with the sacrifice under High Resolution (HR) circumstances. To dealwith the huge domain gap between HR and LR domains and achieve the best on bothdomains, we first took a closer look at the impacts of several resolutionaugmentations and then analyzed the difficulty of LR samples from theperspective of the model gradient produced by different resolution samples.Besides, we also find that the introduction of some resolutions could help thelearning of lower resolutions. Based on these, we divide the LR samples intothree difficulties according to the resolution and propose a more effectiveMulti-Resolution Augmentation. Then, due to the rapidly increasing domain gapas the resolution decreases, we carefully design a novel and effective metricloss based on a LogExp distance function that provides decent gradients toprevent oscillation near the convergence point or tolerance to small distanceerrors; it could also dynamically adjust the penalty for errors in differentdimensions, allowing for more optimization of dimensions with large errors.Combining these two insights, our model could learn more general knowledge in awide resolution range of images and balanced results can be achieved by ourextremely simple framework. Moreover, the augmentations and metrics are thecornerstones of LRFR, so our method could be considered a new baseline for theLRFR task. Experiments on the LRFR datasets: SCface, XQLFW, and large-scaleLRFR dataset: TinyFace demonstrate the effectiveness of our methods, while thedegradation on HRFR datasets is significantly reduced.

Dive into the Resolution Augmentations and Metrics in Low Resolution Face Recognition: A Plain yet Effective New Baseline