CVAIROOct 11, 2024

SmartPretrain: Model-Agnostic and Dataset-Agnostic Representation Learning for Motion Prediction

TsinghuaU of Toronto
arXiv:2410.08669v213 citationsh-index: 44Has CodeICLR
Originality Highly original
AI Analysis

This addresses the data scarcity issue for autonomous vehicle motion prediction, offering a scalable solution that improves performance across datasets.

The paper tackles the problem of limited data for motion prediction in autonomous vehicles by proposing SmartPretrain, a self-supervised learning framework that is model-agnostic and dataset-agnostic, which reduces the MissRate of Forecast-MAE by 10.6%.

Predicting the future motion of surrounding agents is essential for autonomous vehicles (AVs) to operate safely in dynamic, human-robot-mixed environments. However, the scarcity of large-scale driving datasets has hindered the development of robust and generalizable motion prediction models, limiting their ability to capture complex interactions and road geometries. Inspired by recent advances in natural language processing (NLP) and computer vision (CV), self-supervised learning (SSL) has gained significant attention in the motion prediction community for learning rich and transferable scene representations. Nonetheless, existing pre-training methods for motion prediction have largely focused on specific model architectures and single dataset, limiting their scalability and generalizability. To address these challenges, we propose SmartPretrain, a general and scalable SSL framework for motion prediction that is both model-agnostic and dataset-agnostic. Our approach integrates contrastive and reconstructive SSL, leveraging the strengths of both generative and discriminative paradigms to effectively represent spatiotemporal evolution and interactions without imposing architectural constraints. Additionally, SmartPretrain employs a dataset-agnostic scenario sampling strategy that integrates multiple datasets, enhancing data volume, diversity, and robustness. Extensive experiments on multiple datasets demonstrate that SmartPretrain consistently improves the performance of state-of-the-art prediction models across datasets, data splits and main metrics. For instance, SmartPretrain significantly reduces the MissRate of Forecast-MAE by 10.6%. These results highlight SmartPretrain's effectiveness as a unified, scalable solution for motion prediction, breaking free from the limitations of the small-data regime. Codes are available at https://github.com/youngzhou1999/SmartPretrain

Code Implementations1 repo
Foundations

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

Your Notes