Two-Stream Networks for Lane-Change Prediction of Surrounding Vehicles

TL;DR

This paper explores two video-based neural network approaches to predict lane changes of surrounding vehicles within 1-2 seconds, aiming to improve highway safety through early detection of maneuvers.

Contribution

It compares two-stream and spatiotemporal multiplier networks for lane-change prediction using visual cues, analyzing region sizes and prediction horizons.

Findings

Both methods effectively predict lane changes within 1-2 seconds.

Interaction between vehicles and context significantly impacts prediction accuracy.

Different region sizes influence the robustness of the models.

Abstract

In highway scenarios, an alert human driver will typically anticipate early cut-in and cut-out maneuvers of surrounding vehicles using only visual cues. An automated system must anticipate these situations at an early stage too, to increase the safety and the efficiency of its performance. To deal with lane-change recognition and prediction of surrounding vehicles, we pose the problem as an action recognition/prediction problem by stacking visual cues from video cameras. Two video action recognition approaches are analyzed: two-stream convolutional networks and spatiotemporal multiplier networks. Different sizes of the regions around the vehicles are analyzed, evaluating the importance of the interaction between vehicles and the context information in the performance. In addition, different prediction horizons are evaluated. The obtained results demonstrate the potential of these methodologies to serve as robust predictors of future lane-changes of surrounding vehicles in time horizons between 1 and 2 seconds.