Can we learn where people go?

TL;DR

This paper explores using machine learning on video-derived density heatmaps to predict pedestrian destinations at a crossroad, aiming to infer unobservable destination data from sensor inputs.

Contribution

It introduces a novel approach of predicting pedestrian destinations from density heatmaps using Random Forests, providing initial insights into camera placement and data analysis.

Findings

Random Forest can predict destinations from density heatmaps

Camera placement influences prediction accuracy

Proof of concept demonstrated at a crossroad scenario

Abstract

In most agent-based simulators, pedestrians navigate from origins to destinations. Consequently, destinations are essential input parameters to the simulation. While many other relevant parameters as positions, speeds and densities can be obtained from sensors, like cameras, destinations cannot be observed directly. Our research question is: Can we obtain this information from video data using machine learning methods? We use density heatmaps, which indicate the pedestrian density within a given camera cutout, as input to predict the destination distributions. For our proof of concept, we train a Random Forest predictor on an exemplary data set generated with the Vadere microscopic simulator. The scenario is a crossroad where pedestrians can head left, straight or right. In addition, we gain first insights on suitable placement of the camera. The results motivate an in-depth analysis of the methodology.