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

Understanding dynamic hand motions and actions from egocentric RGB videos isa fundamental yet challenging task due to self-occlusion and ambiguity. Toaddress occlusion and ambiguity, we develop a transformer-based framework toexploit temporal information for robust estimation. Noticing the differenttemporal granularity of and the semantic correlation between hand poseestimation and action recognition, we build a network hierarchy with twocascaded transformer encoders, where the first one exploits the short-termtemporal cue for hand pose estimation, and the latter aggregates per-frame poseand object information over a longer time span to recognize the action. Ourapproach achieves competitive results on two first-person hand actionbenchmarks, namely FPHA and H2O. Extensive ablation studies verify our designchoices.

Hierarchical Temporal Transformer for 3D Hand Pose Estimation and Action Recognition from Egocentric RGB Videos