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

While generating realistic body movements, e.g., for avatars in virtualreality, is widely studied in computer vision and graphics, the generation ofeye movements that exhibit realistic coordination with the body remainsunder-explored. We first report a comprehensive analysis of the coordination ofhuman eye and full-body movements during everyday activities based on data fromthe MoGaze and GIMO datasets. We show that eye gaze has strong correlationswith head directions and also full-body motions and there exists a noticeabletime delay between body and eye movements. Inspired by the analyses, we thenpresent Pose2Gaze -- a novel eye-body coordination model that first uses aconvolutional neural network and a spatio-temporal graph convolutional neuralnetwork to extract features from head directions and full-body posesrespectively and then applies a convolutional neural network to generaterealistic eye movements. We compare our method with state-of-the-art methodsthat predict eye gaze only from head movements for three different generationtasks and demonstrate that Pose2Gaze significantly outperforms these baselineson both datasets with an average improvement of 26.4% and 21.6% in mean angularerror, respectively. Our findings underline the significant potential ofcross-modal human gaze behaviour analysis and modelling.

Pose2Gaze: Generating Realistic Human Gaze Behaviour from Full-body Poses using an Eye-body Coordination Model