Abstract
This paper presents a monocular vision based 3D bicycle tracking framework for intelligent vehicles based on a detection method exploiting a deformable part model and a tracking method using an Interacting Multiple Model (IMM) algorithm. Bicycle tracking is important because bicycles share the road with vehicles and can move at comparable speeds in urban environments. From a computer vision standpoint, bicycle detection is challenging as bicycle's appearance can change dramatically between viewpoints and a person riding on the bicycle is a non-rigid object. To this end, we present a tracking-by-detection method to detect and track bicycles that takes into account these difficult issues. First, a mixture model of multiple viewpoints is defined and trained via a Latent Support Vector Machine (LSVM) to detect bicycles under a variety of circumstances. Each model uses a part-based representation. This robust bicycle detector provides a series of measurements (i.e., bounding boxes) in the context of the Kalman filter. Second, to exploit the unique characteristics of bicycle tracking, two motion models based on bicycle's kinematics are fused using an IMM algorithm. For each motion model, an extended Kalman filter (EKF) is used to estimate the position and velocity of a bicycle in the vehicle coordinates. Finally, a single bicycle tracking method using an IMM algorithm is extended to that of multiple bicycle tracking by incorporating a Rao-Blackwellized Particle Filter which runs a particle filter for a data association and an IMM filter for each bicycle tracking. We demonstrate the effectiveness of this approach through a series of experiments run on a new bicycle dataset captured from a vehicle-mounted camera.
Highlights
One of the ultimate goals in the automotive industry is to develop fully autonomous driving vehicles
Another interesting fact is that the error distribution can be well approximated by the Gaussian distribution, which is desirable for the Kalman filter
We quantitatively evaluated our bicycle tracking method using various real world datasets
Summary
One of the ultimate goals in the automotive industry is to develop fully autonomous driving vehicles. One of key subsystems for achieving this goal is a robust perception system that will allow the vehicle to understand its current environment for the safety of people inside and outside of the vehicle [11] Such a perception system must be able to detect and track other traffic participants such as cars as well as the class of objects called vulnerable road users (VRUs) [13] which includes entities such as bicyclists, motorcyclists, and pedestrians. While other technologies such as LIDAR and RADAR are available to autonomous vehicles and all of these different technologies (including various combinations and fusion approaches) have been employed in various ways by other researchers [12], we are primarily considering the challenge of using a vision system to detect and track bicycles.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have