RMP: A Random Mask Pretrain Framework for Motion Prediction

Yi Yang, Qingwen Zhang, Thomas Gilles, Nazre Batool, John Folkesson
9/16/2023
cs.ROcs.CVcs.LG

No official code found. Best implementation: KTH-RPL/RMP (19 stars, last commit 2024-04-18).

Abstract

As the pretraining technique is growing in popularity, little work has been done on pretrained learning-based motion prediction methods in autonomous driving. In this paper, we propose a framework to formalize the pretraining task for trajectory prediction of traffic participants. Within our framework, inspired by the random masked model in natural language processing (NLP) and computer vision (CV), objects' positions at random timesteps are masked and then filled in by the learned neural network (NN). By changing the mask profile, our framework can easily switch among a range of motion-related tasks. We show that our proposed pretraining framework is able to deal with noisy inputs and improves the motion prediction accuracy and miss rate, especially for objects occluded over time by evaluating it on Argoverse and NuScenes datasets.

View on arXivView PDF

Code Implementations(1)

KTH-RPL/RMPCommunity71%

[ITSC'23] Implementation of "RMP: A Random Mask Pretrain Framework for Motion Prediction"

191Jun 5, 20232 years ago

Similar Papers

Traj-MAE: Masked Autoencoders for Trajectory Prediction

86% match2026

SEPT: Towards Efficient Scene Representation Learning for Motion Prediction

85% match2023

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

84% match2 repos2222026

Multi-Transmotion: Pre-trained Model for Human Motion Prediction

83% match1 repo382024

PPT: Pretraining with Pseudo-Labeled Trajectories for Motion Forecasting

83% match2024

Traj-LLM: A New Exploration for Empowering Trajectory Prediction with Pre-trained Large Language Models

83% match2024

Cite this paper

@article{yang2023rmp,
  title  = {RMP: A Random Mask Pretrain Framework for Motion Prediction},
  author = {Yi Yang and Qingwen Zhang and Thomas Gilles and Nazre Batool and John Folkesson},
  year   = {2023},
  eprint = {2309.08989},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2309.08989v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2309.08989.svg)](https://codeofpaper.com/paper/2309.08989)

Discussion