Estimation of Kinematic Motion from Dashcam Footage

Abstract-- The goal of this paper is to explore the accuracy of dashcam footage to predict the actual kinematic motion of a car-like vehicle. Our approach uses ground truth information from the vehicle's on-board data stream, through the controller area network, and a time-synchronized dashboard camera, mounted to a consumer-grade vehicle, for 18 hours of footage and driving. The contributions of the paper include neural network models that allow us to quantify the accuracy of predicting the vehicle speed and yaw, as well as the presence of a lead vehicle, and its relative distance and speed. In addition, the paper describes how other researchers can gather their own data to perform similar experiments, using open-source tools and off-the-shelf technology. One application of the extraction of kinematic data from dashcam videos is the ability to detect specific events that occur while driving from just the footage, such as turns, lane changes, and passes. Others have employed different methods for accomplishing this, such as training videos to descriptions of the videos. However, this relies on having existing descriptions of the events in the videos.