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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Estimating 3D hand pose directly from RGB imagesis challenging but has gainedsteady progress recently bytraining deep models with annotated 3D poses.Howeverannotating 3D poses is difficult and as such only a few 3Dhand posedatasets are available, all with limited samplesizes. In this study, we proposea new framework of training3D pose estimation models from RGB images withoutusingexplicit 3D annotations, i.e., trained with only 2D informa-tion. Ourframework is motivated by two observations: 1)Videos provide richer informationfor estimating 3D posesas opposed to static images; 2) Estimated 3D posesoughtto be consistent whether the videos are viewed in the for-ward order orreverse order. We leverage these two obser-vations to develop a self-supervisedlearning model calledtemporal-aware self-supervised network (TASSN). Byen-forcing temporal consistency constraints, TASSN learns 3Dhand poses andmeshes from videos with only 2D keypointposition annotations. Experiments showthat our modelachieves surprisingly good results, with 3D estimation ac-curacyon par with the state-of-the-art models trained with3D annotations,highlighting the benefit of the temporalconsistency in constraining 3Dprediction models.

Temporal-Aware Self-Supervised Learning for 3D Hand Pose and Mesh Estimation in Videos