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

In this paper, we develop a neural network model to predict future humanmotion from an observed human motion history. We propose a non-autoregressivetransformer architecture to leverage its parallel nature for easier trainingand fast, accurate predictions at test time. The proposed architecture divideshuman motion prediction into two parts: 1) the human trajectory, which is thehip joint 3D position over time and 2) the human pose which is the all otherjoints 3D positions over time with respect to a fixed hip joint. We propose tomake the two predictions simultaneously, as the shared representation canimprove the model performance. Therefore, the model consists of two sets ofencoders and decoders. First, a multi-head attention module applied to encoderoutputs improves human trajectory. Second, another multi-head self-attentionmodule applied to encoder outputs concatenated with decoder outputs facilitateslearning of temporal dependencies. Our model is well-suited for roboticapplications in terms of test accuracy and speed, and compares favorably withrespect to state-of-the-art methods. We demonstrate the real-worldapplicability of our work via the Robot Follow-Ahead task, a challenging yetpractical case study for our proposed model.

STPOTR: Simultaneous Human Trajectory and Pose Prediction Using a Non-Autoregressive Transformer for Robot Following Ahead