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

Detecting facial action units (AU) is one of the fundamental steps inautomatic recognition of facial expression of emotions and cognitive states.Though there have been a variety of approaches proposed for this task, most ofthese models are trained only for the specific target AUs, and as such theyfail to easily adapt to the task of recognition of new AUs (i.e., those notinitially used to train the target models). In this paper, we propose a deeplearning approach for facial AU detection that can easily and in a fast manneradapt to a new AU or target subject by leveraging only a few labeled samplesfrom the new task (either an AU or subject). To this end, we propose a modelingapproach based on the notion of the model-agnostic meta-learning, originallyproposed for the general image recognition/detection tasks (e.g., the characterrecognition from the Omniglot dataset). Specifically, each subject and/or AU istreated as a new learning task and the model learns to adapt based on theknowledge of the previous tasks (the AUs and subjects used to pre-train thetarget models). Thus, given a new subject or AU, this meta-knowledge (that isshared among training and test tasks) is used to adapt the model to the newtask using the notion of deep learning and model-agnostic meta-learning. Weshow on two benchmark datasets (BP4D and DISFA) for facial AU detection thatthe proposed approach can be easily adapted to new tasks (AUs/subjects). Usingonly a few labeled examples from these tasks, the model achieves largeimprovements over the baselines (i.e., non-adapted models).

Fast and Effective Adaptation of Facial Action Unit Detection Deep Model