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

Facial landmark detection aims to localize the anatomically defined points ofhuman faces. In this paper, we study facial landmark detection from partiallylabeled facial images. A typical approach is to (1) train a detector on thelabeled images; (2) generate new training samples using this detector'sprediction as pseudo labels of unlabeled images; (3) retrain the detector onthe labeled samples and partial pseudo labeled samples. In this way, thedetector can learn from both labeled and unlabeled data to become robust. Inthis paper, we propose an interaction mechanism between a teacher and twostudents to generate more reliable pseudo labels for unlabeled data, which arebeneficial to semi-supervised facial landmark detection. Specifically, the twostudents are instantiated as dual detectors. The teacher learns to judge thequality of the pseudo labels generated by the students and filter outunqualified samples before the retraining stage. In this way, the studentdetectors get feedback from their teacher and are retrained by premium datagenerated by itself. Since the two students are trained by different samples, acombination of their predictions will be more robust as the final predictioncompared to either prediction. Extensive experiments on 300-W and AFLWbenchmarks show that the interactions between teacher and students contributeto better utilization of the unlabeled data and achieves state-of-the-artperformance.

Teacher Supervises Students How to Learn From Partially Labeled Images for Facial Landmark Detection