Search

This paper presents a new solution for multi-target tracking over a network of sensors with limited spatial coverage. The proposed solution is based on the centralized data fusion architecture. The main contribution of the paper is the introduction of a new track-to-track fusion approach in which the posterior distributions of multi-target states, reported by various sensor nodes, are fused in a way that the redundant information are combined and the rest complement each other. The proposed solution is formulated within the labeled random finite set framework in which the fused posterior incorporates all the state and label information provided by multiple sensor nodes. The performance of the proposed method is evaluated via simulation experiments that involve challenging tracking scenarios. The proposed method is implemented using sequential Monte Carlo method and the results confirm its effectiveness.

This paper presents an innovative method for motion segmentation in RGB-D dynamic videos with multiple moving objects. The focus is on finding static, small or slow moving objects (often overlooked by other methods) that their inclusion can improve the motion segmentation results. In our approach, semantic object based segmentation and motion cues are combined to estimate the number of moving objects, their motion parameters and perform segmentation. Selective object-based sampling and correspondence matching are used to estimate object specific motion parameters. The main issue with such an approach is the over segmentation of moving parts due to the fact that different objects can have the same motion (e.g. background objects). To resolve this issue, we propose to identify objects with similar motions by characterizing each motion by a distribution of a simple metric and using a statistical inference theory to assess their similarities. To demonstrate the significance of the proposed statistical inference, we present an ablation study, with and without static objects inclusion, on SLAM accuracy using the TUM-RGBD dataset. To test the effectiveness of the proposed method for finding small or slow moving objects, we applied the method to RGB-D MultiBody and SBM-RGBD motion segmentation datasets. The results showed that we can improve the accuracy of motion segmentation for small objects while remaining competitive on overall measures.

A constrained sensor control method is presented for multiobject tracking using labeled multi-Bernoulli filters. The proposed framework is based on a novel approximation of the Cauchy-Schwarz divergence between the labeled multi-Bernoulli prior and posterior densities, which does not need Monte Carlo sampling of random sets in the multiobject space. The void probability functional is also formulated for labeled multi-Bernoulli distributions and used within our proposed method to form a constrained sensor control solution. Numerical studies demonstrate that reasonably acceptable movements are decided for the controlled sensor by our sensor control method, with the advantage that the void probability constraint is formally considered as part of the sensor control optimization algorithm.

In many cases, the multi-target tracking system is essential for realizing the current state of an environment. The standard multi-target tracking algorithms assume that each target state evolves independently and regardless of other targets' states. However, in a real scenario this assumption does not hold in that the motion of any target is dependent on other targets. This paper proposes a new mathematical solution for multi-target tracking system with interacting targets. In the proposed method the prediction operation of the labeled multi-Bernoulli filter is extended to incorporate all possible interactions between targets. The results show that in scenarios where the assumption of a standard motion model is violated, the proposed method achieves higher accuracy for the state estimation of the targets. Also, it shows better performance for estimating the identity of the targets.

In many applications, the states of an unknown number of objects need to be estimated using measurements that are acquired from multiple sensors with different fields of view. When object labels are part of their states, the problem is called the multi-sensor multi-object tracking problem. This paper presents a new solution for statistical fusion of multisensor information in such problems where the sensors form a centralized network. Assuming that a labeled multi-Bernoulli (LMB) filter is running at each sensor node, we suggest a new approach to fuse the multiple LMB posteriors in a centralized manner. The fused posterior is designed to incorporate all the information provided by multiple sensor nodes for each object label. Numerical experiments involving challenging multi-sensor multi-object tracking scenarios show that the proposed method outperforms the state of the art.

In this paper we present a novel non-Bayesian filtering method for tracking multiple objects with a particular application in time-lapse cell microscopic video sequence. In our method the heat-map of the frame sequence is extracted and represented as a pseudo-probability hypothesis density of the image. The pseudo-probability hypothesis density is used as measurements and fused with a prior Poisson random finite set density. We employed Cauchy-Schwarz divergence for information fusion. The presented algorithm was tested on a publicly available cell microscopic video sequence.

Sensor management in multi-object stochastic systems is a theoretically and computationally challenging problem. This paper presents a novel approach to the multi-target multi-sensor control problem within the partially observed Markov decision process (POMDP) framework. We model the multi-object state as a labeled multi-Bernoulli random finite set (RFS), and use the labeled multi-Bernoulli filter in conjunction with minimizing a task-driven control objective function: posterior expected error of cardinality and state (PEECS). A major contribution is a guided search for multi-dimensional optimization in the multi-sensor control command space, using coordinate descent method. In conjunction with the Generalized Covariance Intersection method for multi-sensor fusion, a fast multi-sensor algorithm is achieved. Numerical studies are presented in several scenarios where numerous controllable (mobile) sensors track multiple moving targets with different levels of observability. The results show that our method works significantly faster than the approach taken by a state of art method, with similar tracking errors., 21 pages, 8 figures

This paper presents a track-before-detect labeled multi-Bernoulli filter tailored for industrial mobile platform safety applications. We derive two application specific separable likelihood functions that capture the geometric shape and colour information of the human targets who are wearing a high visible vest. These likelihoods are then used in a labeled multi-Bernoulli filter with a novel two step Bayesian update. Preliminary simulation results show that the proposed solution can successfully track human workers wearing a luminous yellow colour vest in an industrial environment., The conference which this paper was submitted, has rejected this paper. Thus, we are in the process of enhancing the content of the paper and submit it to another conference/journal