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

WiFi-based pose estimation is a technology with great potential for thedevelopment of smart homes and metaverse avatar generation. However, currentWiFi-based pose estimation methods are predominantly evaluated under controlledlaboratory conditions with sophisticated vision models to acquire accuratelylabeled data. Furthermore, WiFi CSI is highly sensitive to environmentalvariables, and direct application of a pre-trained model to a new environmentmay yield suboptimal results due to domain shift. In this paper, we proposes adomain adaptation algorithm, AdaPose, designed specifically forweakly-supervised WiFi-based pose estimation. The proposed method aims toidentify consistent human poses that are highly resistant to environmentaldynamics. To achieve this goal, we introduce a Mapping Consistency Loss thataligns the domain discrepancy of source and target domains based on innerconsistency between input and output at the mapping level. We conduct extensiveexperiments on domain adaptation in two different scenes using ourself-collected pose estimation dataset containing WiFi CSI frames. The resultsdemonstrate the effectiveness and robustness of AdaPose in eliminating domainshift, thereby facilitating the widespread application of WiFi-based poseestimation in smart cities.

AdaPose: Towards Cross-Site Device-Free Human Pose Estimation with Commodity WiFi