Estimation of Kinematic Motion from Dashcam Footage
Evelyn Zhang, Alex Richardson, Jonathan Sprinkle

TL;DR
This paper investigates how accurately dashcam footage can be used to estimate a vehicle's kinematic motion by combining neural networks with on-board data, providing a methodology for data collection and analysis.
Contribution
It introduces neural network models for predicting vehicle speed, yaw, and lead vehicle information from dashcam footage, and details a reproducible data collection process using accessible tools.
Findings
Neural networks can accurately predict vehicle speed and yaw from dashcam footage.
The approach effectively detects lead vehicles and estimates their relative distance and speed.
Open-source tools enable others to replicate and extend the experiments.
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.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsVehicle Dynamics and Control Systems · Autonomous Vehicle Technology and Safety · Aerospace and Aviation Technology
