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

Nowadays, deploying a robust face recognition product becomes easy with thedevelopment of face recognition techniques for decades. Not only profile imageverification but also the state-of-the-art method can handle the in-the-wildimage almost perfectly. However, the concern of privacy issues raise rapidlysince mainstream research results are powered by tons of web-crawled data,which faces the privacy invasion issue. The community tries to escape thispredicament completely by training the face recognition model with syntheticdata but faces severe domain gap issues, which still need to access real imagesand identity labels to fine-tune the model. In this paper, we propose SASMU, asimple, novel, and effective method for face recognition using a syntheticdataset. Our proposed method consists of spatial data augmentation (SA) andspectrum mixup (SMU). We first analyze the existing synthetic datasets fordeveloping a face recognition system. Then, we reveal that heavy dataaugmentation is helpful for boosting performance when using synthetic data. Byanalyzing the previous frequency mixup studies, we proposed a novel method fordomain generalization. Extensive experimental results have demonstrated theeffectiveness of SASMU, achieving state-of-the-art performance on severalcommon benchmarks, such as LFW, AgeDB-30, CA-LFW, CFP-FP, and CP-LFW.

SASMU: boost the performance of generalized recognition model using synthetic face dataset