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 great success of Convolutional Neural Networks (CNN) for facial attributeprediction relies on a large amount of labeled images. Facial image datasetsare usually annotated by some commonly used attributes (e.g., gender), whilelabels for the other attributes (e.g., big nose) are limited which causes theirprediction challenging. To address this problem, we use a new Multi-TaskLearning (MTL) paradigm in which a facial attribute predictor uses theknowledge of other related attributes to obtain a better generalizationperformance. Here, we leverage MLT paradigm in two problem settings. First, itis assumed that the structure of the tasks (e.g., grouping pattern of facialattributes) is known as a prior knowledge, and parameters of the tasks (i.e.,predictors) within the same group are represented by a linear combination of alimited number of underlying basis tasks. Here, a sparsity constraint on thecoefficients of this linear combination is also considered such that each taskis represented in a more structured and simpler manner. Second, it is assumedthat the structure of the tasks is unknown, and then structure and parametersof the tasks are learned jointly by using a Laplacian regularization framework.Our MTL methods are compared with competing methods for facial attributeprediction to show its effectiveness.

Tasks Structure Regularization in Multi-Task Learning for Improving Facial Attribute Prediction