CVAIJun 30, 2022

TENET: Transformer Encoding Network for Effective Temporal Flow on Motion Prediction

arXiv:2207.00170v116 citationsh-index: 11
Originality Highly original
AI Analysis

This work addresses motion forecasting for autonomous vehicles, representing an incremental improvement with a novel method for a known bottleneck.

The authors tackled motion prediction for autonomous driving by developing a Transformer-based method with a Temporal Flow Header and K-means ensemble, achieving first place in the Argoverse 2 Motion Forecasting Challenge with a state-of-the-art brier-minFDE score of 1.90.

This technical report presents an effective method for motion prediction in autonomous driving. We develop a Transformer-based method for input encoding and trajectory prediction. Besides, we propose the Temporal Flow Header to enhance the trajectory encoding. In the end, an efficient K-means ensemble method is used. Using our Transformer network and ensemble method, we win the first place of Argoverse 2 Motion Forecasting Challenge with the state-of-the-art brier-minFDE score of 1.90.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes