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

The ability to sense, localize, and estimate the 3D position and orientationof the human body is critical in virtual reality (VR) and extended reality (XR)applications. This becomes more important and challenging with the deploymentof VR/XR applications over the next generation of wireless systems such as 5Gand beyond. In this paper, we propose a novel framework that can reconstructthe 3D human body pose of the user given sparse measurements from InertialMeasurement Unit (IMU) sensors over a noisy wireless environment. Specifically,our framework enables reliable transmission of compressed IMU signals throughnoisy wireless channels and effective recovery of such signals at the receiver,e.g., an edge server. This task is very challenging due to the constraints oftransmit power, recovery accuracy, and recovery latency. To address thesechallenges, we first develop a deep generative model at the receiver to recoverthe data from linear measurements of IMU signals. The linear measurements ofthe IMU signals are obtained by a linear projection with a measurement matrixbased on the compressive sensing theory. The key to the success of ourframework lies in the novel design of the measurement matrix at thetransmitter, which can not only satisfy power constraints for the IMU devicesbut also obtain a highly accurate recovery for the IMU signals at the receiver.This can be achieved by extending the set-restricted eigenvalue condition ofthe measurement matrix and combining it with an upper bound for the powertransmission constraint. Our framework can achieve robust performance forrecovering 3D human poses from noisy compressed IMU signals. Additionally, ourpre-trained deep generative model achieves signal reconstruction accuracycomparable to an optimization-based approach, i.e., Lasso, but is an order ofmagnitude faster.

Reconstructing Human Pose from Inertial Measurements: A Generative Model-based Compressive Sensing Approach