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

06/30/2022
by   Yuting Wang, et al.
0

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.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
07/13/2022

QML for Argoverse 2 Motion Forecasting Challenge

To safely navigate in various complex traffic scenarios, autonomous driv...
research
03/08/2022

Occupancy Flow Fields for Motion Forecasting in Autonomous Driving

We propose Occupancy Flow Fields, a new representation for motion foreca...
research
02/16/2023

Robust Human Motion Forecasting using Transformer-based Model

Comprehending human motion is a fundamental challenge for developing Hum...
research
02/10/2023

Data-Driven Stochastic Motion Evaluation and Optimization with Image by Spatially-Aligned Temporal Encoding

This paper proposes a probabilistic motion prediction method for long mo...
research
06/25/2022

SC-Transformer++: Structured Context Transformer for Generic Event Boundary Detection

This report presents the algorithm used in the submission of Generic Eve...
research
06/28/2023

The 2nd Place Solution for 2023 Waymo Open Sim Agents Challenge

In this technical report, we present the 2nd place solution of 2023 Waym...
research
07/02/2022

Golfer: Trajectory Prediction with Masked Goal Conditioning MnM Network

Transformers have enabled breakthroughs in NLP and computer vision, and ...

Please sign up or login with your details

Forgot password? Click here to reset