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 number of the training samples per person has a significant impact on face recognition (FR) performance. For the single training sample per person (STSPP) problem, most traditional FR algorithms exhibit performance degradation owing to the limited information available to predict the variance of the query sample. This paper proposes a new method for the STSPP problem in FR, namely the Learn-Generate-Classify (LGC) method. The LGC method first learns the relationship between the multiple images of a subject based on dictionary learning from a generic training set. Then it predicts the intra-class variance of the gallery set using the learned relationship. Based on the predicted information, synthetic samples can be generated, thus extending the single sample gallery set to one having multiple samples. Finally, we can classify the query samples using the traditional sparse representation classification (SRC) framework on the multisample gallery set. We verified the effectiveness of the new LGC method on the CMU Multi-pie database, with different illumination, expression and pose variation factors. It shows that the LGC method demonstrates a promising FR performance with only a STSPP.

A New Face Recognition Algorithm based on Dictionary Learning for a Single Training Sample per Person