Multiple Target Tracking Applications Research Articles

Image-Based Multi-Target Tracking through Multi-Bernoulli Filtering with Interactive Likelihoods.

We develop an interactive likelihood (ILH) for sequential Monte Carlo (SMC) methods for image-based multiple target tracking applications. The purpose of the ILH is to improve tracking accuracy by reducing the need for data association. In addition, we integrate a recently developed deep neural network for pedestrian detection along with the ILH with a multi-Bernoulli filter. We evaluate the performance of the multi-Bernoulli filter with the ILH and the pedestrian detector in a number of publicly available datasets (2003 PETS INMOVE, Australian Rules Football League (AFL) and TUD-Stadtmitte) using standard, well-known multi-target tracking metrics (optimal sub-pattern assignment (OSPA) and classification of events, activities and relationships for multi-object trackers (CLEAR MOT)). In all datasets, the ILH term increases the tracking accuracy of the multi-Bernoulli filter.

An Efficient MHT Implementation Using GRASP

The multiple hypothesis tracking (MHT) approach has been proven to be successful in multiple target tracking applications, however, its computational complexity remains a major hurdle to its practical implementation. This paper presents an efficient MHT implementation, referred to as “GRASP-MHT”, which integrates a greedy randomized adaptive search procedure (GRASP) within a track-oriented MHT framework. The hypothesis generating problem arising in the MHT framework is formulated as a maximum weighted independent set problem, and a GRASP module is designed to generate multiple high-quality hypotheses. An extensive simulation study was carried out to compare the performance of the proposed GRASP-MHT against several well-known multitarget tracking algorithms, and multiple metrics were considered in order to make the performance evaluation more comprehensive. Experimental results indicate that, by efficiently generating and fusing multiple high-quality global hypotheses in the data association process, GRASP-MHT is able to achieve better overall tracking performance than other algorithms, especially in a closely-spaced multitarget scenario.

Sensor management using an active sensing approach

An approach that is common in the machine learning literature, known as active sensing, is applied to provide a method for managing agile sensors in a dynamic environment. We adopt an active sensing approach to scheduling sensors for multiple target tracking applications that combines particle filtering, predictive density estimation, and relative entropy maximization. Specifically, the goal of the system is to learn the number and states of a group of moving targets occupying a surveillance region. At each time step, the system computes a sensing action to take, based on an entropy measure called the Rényi divergence. After the measurement is made, the system updates its probability density on the number and states of the targets. This procedure repeats at each time where a sensor is available for use. The algorithms developed here extend standard active sensing methodology to dynamically evolving objects and continuous state spaces of high dimension. It is shown using simulated measurements on real recorded target trajectories that this method of sensor management yields more than a ten fold gain in sensor efficiency when compared to periodic scanning.

A New Lagrangian Relaxation Based Algorithm for a Class ofMultidimensional Assignment Problems

Large classes of data association problems in multiple target tracking applications involving both multiple and single sensor systems can be formulated as multidimensional assignment problems. These NP-hard problems are large scale and sparse with noisy objective function values, but must be solved in “real-time”. Lagrangian relaxation methods have proven to be particularly effective in solving these problems to the noise level in real-time, especially for dense scenarios and for multiple scans of data from multiple sensors. This work presents a new class of constructive Lagrangian relaxation algorithms that circumvent some of the deficiencies of previous methods. The results of several numerical studies demonstrate the efficiency and effectiveness of the new algorithm class.