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

Forecasting short-term motion of nearby vehicles presents an inherentlychallenging issue as the space of their possible future movements is notstrictly limited to a set of single trajectories. Recently proposed techniquesthat demonstrate plausible results concentrate primarily on forecasting a fixednumber of deterministic predictions, or on classifying over a wide variety oftrajectories that were previously generated using e.g. dynamic model. Thispaper focuses on addressing the uncertainty associated with the discussed taskby utilising the stochastic nature of generative models in order to produce adiverse set of plausible paths with regards to tracked vehicles. Morespecifically, we propose to account for the multi-modality of the problem withuse of Conditional Variational Autoencoder (C-VAE) conditioned on an agent'spast motion as well as a rasterised scene context encoded with Capsule Network(CapsNet). In addition, we demonstrate advantages of employing the Minimum overN (MoN) cost function which measures the distance between ground truth and Ngenerated samples and tries to minimise the loss with respect to the closestsample, effectively leading to more diverse predictions. We examine our networkon a publicly available dataset against recent state-of-the-art methods andshow that our approach outperforms these techniques in numerous scenarioswhilst significantly reducing the number of trainable parameters as well asallowing to sample an arbitrary amount of diverse trajectories.

Multi-modal anticipation of stochastic trajectories in a dynamic environment with Conditional Variational Autoencoders