3-D Segmentation and Motion Estimation of Range Data for Crash Prevention
In this paper we present a 3-D object segmentation and motion estimation scheme of range video data for crash prevention applications. The segmentation is based on slope values of every point in the scene with which atomic regions are constructed by region growing. Atomic regions are merged with the flat surfaces (non-ground points) according to their mean height and the range difference of neighboring points. Following segmentation we apply a 3-D recursive motion estimation algorithm to each moving object. Simulation results show that the segmentation scheme together with the recursive motion estimation algorithm can be highly effective in estimating 3-D motion parameters of multiple moving objects in range video data.
3-D Segmentation and Motion Estimation of Range Data for Crash Prevention, NIST Interagency/Internal Report (NISTIR), National Institute of Standards and Technology, Gaithersburg, MD, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=51181
(Accessed March 3, 2024)