XVTP3D: Cross-view Trajectory Prediction Using Shared 3D Queries for Autonomous Driving

TL;DR

This paper introduces XVTP3D, a novel cross-view trajectory prediction method for autonomous driving that uses shared 3D queries to ensure consistency across multiple sensor views, improving prediction accuracy.

Contribution

It presents a new approach employing shared 3D queries and a top-down paradigm for trajectory prediction, addressing cross-view consistency issues.

Findings

Achieved state-of-the-art performance on two datasets.

Maintained cross-view consistency in predictions.

Enhanced robustness with novel feature capturing methods.

Abstract

Trajectory prediction with uncertainty is a critical and challenging task for autonomous driving. Nowadays, we can easily access sensor data represented in multiple views. However, cross-view consistency has not been evaluated by the existing models, which might lead to divergences between the multimodal predictions from different views. It is not practical and effective when the network does not comprehend the 3D scene, which could cause the downstream module in a dilemma. Instead, we predicts multimodal trajectories while maintaining cross-view consistency. We presented a cross-view trajectory prediction method using shared 3D Queries (XVTP3D). We employ a set of 3D queries shared across views to generate multi-goals that are cross-view consistent. We also proposed a random mask method and coarse-to-fine cross-attention to capture robust cross-view features. As far as we know, this is the first work that introduces the outstanding top-down paradigm in BEV detection field to a trajectory prediction problem. The results of experiments on two publicly available datasets show that XVTP3D achieved state-of-the-art performance with consistent cross-view predictions.