Automatic creation of urban velocity fields from aerial video

TL;DR

This paper introduces a system that models vehicle motion in urban scenes from aerial videos by creating pixel-wise velocity distributions, enabling analysis of traffic patterns and environmental traffic rules.

Contribution

The paper presents a complete system for extracting and modeling vehicle velocities from low frame-rate aerial videos, integrating stabilization, segmentation, tracking, and probabilistic velocity modeling.

Findings

Velocity distributions reveal lane directions and speed limits.

The system accurately tracks vehicle movements over time.

Traffic bottlenecks and common trajectories are identifiable.

Abstract

In this paper, we present a system for modelling vehicle motion in an urban scene from low frame-rate aerial video. In particular, the scene is modelled as a probability distribution over velocities at every pixel in the image. We describe the complete system for acquiring this model. The video is captured from a helicopter and stabilized by warping the images to match an orthorectified image of the area. A pixel classifier is applied to the stabilized images, and the response is segmented to determine car locations and orientations. The results are fed in to a tracking scheme which tracks cars for three frames, creating tracklets. This allows the tracker to use a combination of velocity, direction, appearance, and acceleration cues to keep only tracks likely to be correct. Each tracklet provides a measurement of the car velocity at every point along the tracklet's length, and these are then aggregated to create a histogram of vehicle velocities at every pixel in the image. The results demonstrate that the velocity probability distribution prior can be used to infer a variety of information about road lane directions, speed limits, vehicle speeds and common trajectories, and traffic bottlenecks, as well as providing a means of describing environmental knowledge about traffic rules that can be used in tracking.