The authors propose a visual-inertial approach to estimate the body-fixed lateral velocity of motorcycles traveling along extra-urban roads. The approach comprises the following steps: First, a monocular camera takes video of the road ahead. Key features from sequential images of the road surface are extracted using the Harris corner detector and matching features are identified using the Fast retina keypoint descriptor. The locations of these features on the road surface are determined using a mapping based on an intuitive ray-casting approach. Next, the feature locations on the road, the angular velocity measurements and the optical flow of the feature projection locations on the image plane are used to formulate the egomotion of the motorcycle as a system of linear equations from which a velocity estimate is solved for using the least-squares method. Finally, this estimate is fused with readings from an inertial navigation system using a Kalman filter to produce a filtered estimate and correct integrator drift. The approach is validated against simulation data generated using BikeSim and the results are compared against state observer approaches and previously published visual-inertial approaches from the authors.