RNN-based Pedestrian Crossing Prediction using Activity and Pose-related Features

TL;DR

This paper proposes deep learning models combining CNN and RNN to predict pedestrian crossing intentions, emphasizing feature extraction methods and additional variables like gaze direction to improve accuracy in autonomous driving scenarios.

Contribution

It introduces various RNN-based models with different feature extraction techniques and additional variables, evaluated on the JAAD dataset for pedestrian crossing prediction.

Findings

Feature extraction method significantly affects performance

Including gaze direction improves prediction accuracy

Model choice impacts overall effectiveness

Abstract

Pedestrian crossing prediction is a crucial task for autonomous driving. Numerous studies show that an early estimation of the pedestrian's intention can decrease or even avoid a high percentage of accidents. In this paper, different variations of a deep learning system are proposed to attempt to solve this problem. The proposed models are composed of two parts: a CNN-based feature extractor and an RNN module. All the models were trained and tested on the JAAD dataset. The results obtained indicate that the choice of the features extraction method, the inclusion of additional variables such as pedestrian gaze direction and discrete orientation, and the chosen RNN type have a significant impact on the final performance.