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

Given the similarity between facial expression categories, the presence ofcompound facial expressions, and the subjectivity of annotators, facialexpression recognition (FER) datasets often suffer from ambiguity and noisylabels. Ambiguous expressions are challenging to differentiate from expressionswith noisy labels, which hurt the robustness of FER models. Furthermore, thedifficulty of recognition varies across different expression categories,rendering a uniform approach unfair for all expressions. In this paper, weintroduce a novel approach called Adaptive Sample Mining (ASM) to dynamicallyaddress ambiguity and noise within each expression category. First, theAdaptive Threshold Learning module generates two thresholds, namely the cleanand noisy thresholds, for each category. These thresholds are based on the meanclass probabilities at each training epoch. Next, the Sample Mining modulepartitions the dataset into three subsets: clean, ambiguity, and noise, bycomparing the sample confidence with the clean and noisy thresholds. Finally,the Tri-Regularization module employs a mutual learning strategy for theambiguity subset to enhance discrimination ability, and an unsupervisedlearning strategy for the noise subset to mitigate the impact of noisy labels.Extensive experiments prove that our method can effectively mine both ambiguityand noise, and outperform SOTA methods on both synthetic noisy and originaldatasets. The supplement material is available athttps://github.com/yang/ASM.

ASM: Adaptive Sample Mining for In-The-Wild Facial Expression Recognition