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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Face verification can be regarded as a 2-class fine-grained visualrecognition problem. Enhancing the feature's discriminative power is one of thekey problems to improve its performance. Metric learning technology is oftenapplied to address this need, while achieving a good tradeoff betweenunderfitting and overfitting plays the vital role in metric learning. Hence, wepropose a novel ensemble cascade metric learning (ECML) mechanism. Inparticular, hierarchical metric learning is executed in the cascade way toalleviate underfitting. Meanwhile, at each learning level, the features aresplit into non-overlapping groups. Then, metric learning is executed among thefeature groups in the ensemble manner to resist overfitting. Considering thefeature distribution characteristics of faces, a robust Mahalanobis metriclearning method (RMML) with closed-form solution is additionally proposed. Itcan avoid the computation failure issue on inverse matrix faced by somewell-known metric learning approaches (e.g., KISSME). Embedding RMML into theproposed ECML mechanism, our metric learning paradigm (EC-RMML) can run in theone-pass learning manner. Experimental results demonstrate that EC-RMML issuperior to state-of-the-art metric learning methods for face verification.And, the proposed ensemble cascade metric learning mechanism is also applicableto other metric learning approaches.

ECML: An Ensemble Cascade Metric Learning Mechanism towards Face Verification