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

The cosine-based softmax losses and their variants achieve great success indeep learning based face recognition. However, hyperparameter settings in theselosses have significant influences on the optimization path as well as thefinal recognition performance. Manually tuning those hyperparameters heavilyrelies on user experience and requires many training tricks. In this paper, weinvestigate in depth the effects of two important hyperparameters ofcosine-based softmax losses, the scale parameter and angular margin parameter,by analyzing how they modulate the predicted classification probability. Basedon these analysis, we propose a novel cosine-based softmax loss, AdaCos, whichis hyperparameter-free and leverages an adaptive scale parameter toautomatically strengthen the training supervisions during the training process.We apply the proposed AdaCos loss to large-scale face verification andidentification datasets, including LFW, MegaFace, and IJB-C 1:1 Verification.Our results show that training deep neural networks with the AdaCos loss isstable and able to achieve high face recognition accuracy. Our methodoutperforms state-of-the-art softmax losses on all the three datasets.

AdaCos: Adaptively Scaling Cosine Logits for Effectively Learning Deep Face Representations