Object level footprint uncertainty quantification in infrastructure based sensing

TL;DR

This paper develops a method to quantify object footprint uncertainty in infrastructure-based camera sensing by relating camera errors to ground coordinates and validating with LiDAR data and simulation.

Contribution

It introduces a closed-form relationship between camera errors and ground coordinates, enabling footprint uncertainty quantification in infrastructure sensing.

Findings

Uncertainty can be modeled as a function of camera errors and object extremities.

A method to estimate typical camera error sizes using LiDAR ground truth.

Simulation demonstrates how uncertainty varies during a vehicle maneuver.

Abstract

We examine the problem of estimating footprint uncertainty of objects imaged using the infrastructure based camera sensing. A closed form relationship is established between the ground coordinates and the sources of the camera errors. Using the error propagation equation, the covariance of a given ground coordinate can be measured as a function of the camera errors. The uncertainty of the footprint of the bounding box can then be given as the function of all the extreme points of the object footprint. In order to calculate the uncertainty of a ground point, the typical error sizes of the error sources are required. We present a method of estimating the typical error sizes from an experiment using a static, high-precision LiDAR as the ground truth. Finally, we present a simulated case study of uncertainty quantification from infrastructure based camera in CARLA to provide a sense of how the uncertainty changes across a left turn maneuver.