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

This study explores the application of self-supervised learning (SSL) to thetask of motion forecasting, an area that has not yet been extensivelyinvestigated despite the widespread success of SSL in computer vision andnatural language processing. To address this gap, we introduce Forecast-MAE, anextension of the mask autoencoders framework that is specifically designed forself-supervised learning of the motion forecasting task. Our approach includesa novel masking strategy that leverages the strong interconnections betweenagents' trajectories and road networks, involving complementary masking ofagents' future or history trajectories and random masking of lane segments. Ourexperiments on the challenging Argoverse 2 motion forecasting benchmark showthat Forecast-MAE, which utilizes standard Transformer blocks with minimalinductive bias, achieves competitive performance compared to state-of-the-artmethods that rely on supervised learning and sophisticated designs. Moreover,it outperforms the previous self-supervised learning method by a significantmargin. Code is available at https://github.com/jchengai/forecast-mae.

Forecast-MAE: Self-supervised Pre-training for Motion Forecasting with Masked Autoencoders