From Pedestrian Detection to Crosswalk Estimation: An EM Algorithm and Analysis on Diverse Datasets
This work addresses the specific problem of infrastructure mapping for autonomous vehicles and urban planning by providing a practical crosswalk estimation method.
The paper tackles the problem of estimating pedestrian crosswalk geometry from pedestrian detection data, presenting an EM algorithm that achieves accurate corner point and crossing segment estimation for both marked and unmarked crosswalks across three diverse real-world datasets.
In this work, we contribute an EM algorithm for estimation of corner points and linear crossing segments for both marked and unmarked pedestrian crosswalks using the detections of pedestrians from processed LiDAR point clouds or camera images. We demonstrate the algorithmic performance by analyzing three real-world datasets containing multiple periods of data collection for four-corner and two-corner intersections with marked and unmarked crosswalks. Additionally, we include a Python video tool to visualize the crossing parameter estimation, pedestrian trajectories, and phase intervals in our public source code.