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 Kalman filter (KF) is a widely-used algorithm for tracking dynamicsystems that are captured by state space (SS) models. The need to fullydescribe a SS model limits its applicability under complex settings, e.g., whentracking based on visual data, and the processing of high-dimensional signalsoften induces notable latency. These challenges can be treated by mapping themeasurements into latent features obeying some postulated closed-form SS model,and applying the KF in the latent space. However, the validity of thisapproximated SS model may constitute a limiting factor. In this work, we studytracking from high-dimensional measurements under complex settings using ahybrid model-based/data-driven approach. By gradually tackling the challengesin handling the observations model and the task, we develop Latent-KalmanNet,which implements tracking from high-dimensional measurements by leveraging datato jointly learn the KF along with the latent space mapping. Latent-KalmanNetcombines a learned encoder with data-driven tracking in the latent space usingthe recently proposed-KalmanNet, while identifying the ability of each of thesetrainable modules to assist its counterpart via providing a suitable prior (byKalmanNet) and by learning a latent representation that facilitates data-aidedtracking (by the encoder). Our empirical results demonstrate that the proposedLatent-KalmanNet achieves improved accuracy and run-time performance over bothmodel-based and data-driven techniques by learning a surrogate latentrepresentation that most facilitates tracking, while operating with limitedcomplexity and latency.

Latent-KalmanNet: Learned Kalman Filtering for Tracking from High-Dimensional Signals