The Effect of Tracker and Control Point Distributions on Vehicle Position and Speed Estimates from Dash Camera Video

Dash cameras (dashcams) can provide collision reconstructionists with quantifiable vehicle position and speed estimates. These estimates are achieved by tracking 2D video features with camera-tracking software to solve for the camera position history, and speed can then be calculated from the position-time history. Not all scenes have the same geometric features in quality or abundance. In this study, we compared the vehicle position and derived-speed estimates from dashcam video for different numbers and spatial distributions of tracked features that mimicked the continuum between barren environments and feature-rich environments. We used video from a dashcam mounted in a vehicle undergoing straight-line emergency braking. The surrounding environment had abundant trackable features on both sides of the road, including road markings, streetlights, signs, trees, and buildings. We first created a reference solution using SynthEyes, a 3D camera- and object-tracking program, and all of the available features. We then created eight test conditions by eliminating the available 2D trackers and 3D control points in large volumetric regions to mimic less feature-rich environments. To compare these different conditions, we calculated the bias, uncertainty and mean absolute error in the camera position and speed estimates between the test conditions and the reference condition. For six of the eight test conditions, we found that removing large regions of the available tracker points and control points generated camera solutions with relatively small position errors (mean absolute errors of 32 to 119 mm) and small speed errors (<1 km/h) compared to the reference solution. One test condition did not reach a solution and another test condition with a poor distribution of few control points had larger errors. Overall, these findings showed that the camera-tracking methodology used here for calculating a dashcam's position and speed remained robust even in some relatively feature-poor environments.