Estimating initial pose by utilizing symmetric property for real-time intelligent transportation system

Abstract

Accurate initial pose is a kind of crucial information when localizing shape features or recognizing car models in a real-time intelligent transportation system. However, obtaining a sufficiently accurate initial pose is generally difficult because very little prior information is available when the vehicle is first time captured by the camera. This paper proposes an algorithm that can provide a sufficiently accurate initial pose estimation without incurring a heavy computation load. This algorithm locates the symmetric center of cars and utilizes it to estimate the initial pose of vehicles. Theoretical analysis and real data experiments confirm that the proposed algorithm provides significantly better accuracy in initial pose estimation than previous pose estimation methods.