Abstract—Occlusion and lack of visibility in crowded and cluttered scenes make it difficult to track individual people correctly and consistently, particularly in a single view. We present a multiview approach to solve this problem. In our approach, we neither detect nor track objects from any single camera or camera pair; rather, evidence is gathered from all of the cameras into a synergistic framework and detection and tracking results are propagated back to each view. Unlike other multiview approaches that require fully calibrated views, our approach is purely image-based and uses only 2D constructs. To this end, we develop a planar homographic occupancy constraint that fuses foreground likelihood information from multiple views to resolve occlusions and localize people on a reference scene plane. For greater robustness, this process is extended to multiple planes parallel to the reference plane in the framework of plane to plane homologies. Our fusion methodology also models scene clutter using the Schmieder and Weathersby clutter measure, which acts as a confidence prior, to assign higher fusion weight to views with lesser clutter. Detection and tracking are performed simultaneously by graph cuts segmentation of tracks in the space-time occupancy likelihood data. Experimental results with detailed qualitative and quantitative analysis are demonstrated in challenging multiview crowded scenes. Index Terms—Tracking, sensor fusion, graph-theoretic methods. Ç 1

Development of multiple camera based vision systems for analysis of dynamic objects such as humans is challenging due to occlusions and similarity in the appearance of a person with the background and other people- visual “confusion”. Since occlusion and confusion depends on the presence of other people in the scene, it leads to a dependency structure where there are often loops in the resulting Bayesian network. While approaches such as loopy belief propagation can be used for inference, they are computationally expensive and convergence is not guaranteed in many situations. We present a unified approach, COST, that reasons about such dependencies and yields an order for the inference of each person in a group of people and a set of cameras to be used for inferences for a person. Using the probabilistic distribution of the positions and appearances of people, COST performs visibility and confusion analysis for each part of each person and computes the amount of information that can be computed with and without more accurate estimation of the positions of other people. We present an optimization problem to select set of cameras and inference dependencies for each person which attempts to minimize the computational cost under given performance constraints. Results show the efficiency of COST in improving the performance of such systems and reducing the computational resources required. 1.

Several interesting monitoring applications concern people entering a prescribed area, where they deposit an object in their possession, or collect an object deposited earlier. One example arises in the use of bicycle racks. We propose a novel method for associating each person who deposits an object with the person who later collects it. Our main contribution is to deal with ambiguity in the visual data through the use of global constraints on what is possible. The method is evaluated on a set of practical experiments in a bicycle rack, and applied to online theft detection by comparing the colour profile of associated individuals. 1

In this paper we present a novel vision-based system for the automated production of personalised video souvenirs for visitors in leisure and cultural heritage venues. Visitors are visually identified and tracked through a camera network. The system produces a personalized DVD souvenir at the end of a visitor’s stay allowing visitors to relive their experiences. We analyze how we identify visitors by fusing facial and body features, how we track visitors, how the tracker recovers from failures due to occlusions, as well as how we annotate and compile the final product. Our experiments demonstrate the feasibility of the proposed approach. Key words: human identification, tracking, automated content production 1

In a crowded public space, people often walk in groups, either with people they know or strangers. Associating a group of people over space and time can assist understanding individual’s behaviours as it provides vital visual context for matching individuals within the group. Seemingly an ‘easier ’ task compared with person matching given more and richer visual content, this problem is in fact very challenging because a group of people can be highly non-rigid with changing relative position of people within the group and severe self-occlusions. In this paper, for the first time, the problem of matching/associating groups of people over large space and time captured in multiple non-overlapping camera views is addressed. Specifically, a novel people group representation and a group matching algorithm are proposed. The former addresses changes in the relative positions of people in a group and the latter deals with variations in illumination and viewpoint across camera views. In addition, we demonstrate a notable enhancement on individual person matching by utilising the group description as visual context. Our methods are validated using the 2008 i-LIDS Multiple-Camera Tracking Scenario (MCTS) dataset on multiple camera views from a busy airport arrival hall. 1

In this work a bottom-up approach for human behaviour understanding is presented, using a multi-camera system. The proposed methodology classifies behaviour as normal or abnormal, by treating short-term behaviour classification and trajectory classification as two different classification problems. Based on that assumption, a set of calculated features provide input to two one-class classifiers: a Support Vector Machine and a continuous Hidden Markov Model treated as an one-class classifier. An approximation algorithm, referring to the Forward Backward procedure of the continuous Hidden Markov Model, is also proposed to overcome numerical stability problems in the calculation of probability of emission for very long observations.

Abstract. This paper presents an automatic visual analysis system for simultaneously tracking humans and vehicles using multiple cameras in an unconstrained outdoor environment. The system establishes correspondence between views using a principal axis approach for humans and a footage region approach for vehicles. Novel methods for locating humans in groups and solving ambiguity when matching vehicles across views are presented. Foreground segmentation for each view is performed using the codebook method and HSV shadow suppression. The tracking of objects is performed in each view, and occlusion situations are resolved by probabilistic appearance models. The system is tested on hours of video and on three different datasets. 1

Multiple people tracking from multiple cameras benefits many applications in computer vision. However, using the existing methods, various problems such as inter-person occlusions still degrade the position estimations. In this paper, we attempt to solve the problems by analyzing the view visibility and ranking the reliability of the cues from 2D views. We combine the visibility with the smoothness constraints into a probability framework, which offers a more flexible and robust estimation. Aside from that, in this paper, we also introduce 2D reference lines to estimate the 2D position of every person in the input images. These lines are able to estimate more accurate and robust 2D positions. We quantitatively evaluate our method by using both our own multiple-people data set and a public data set. The evaluation and experimental results on the standard data set show that our methods considerably improve the accuracy and the robustness. 1

Abstract. In the paper, we proposed a system for automatic detection and tracking of multiple targets in a multi–camera surveillance zone. In each camera view of this system, we only need a simple object detection algorithm, such as background subtraction. The detection results from multiple cameras are fused into a posterior distribution, named TDP, based on the Bayesian rule. This TDP distribution indicates the likelihood of having some moving elements on the ground plane. To properly handle the tracking of multiple moving targets over time, a sample–based framework, which combines Markov Chain Monte Carlo (MCMC), Sequential Monte Carlo (SMC), and Mean–Shift clustering, is proposed. The MCMC is used to handle the occurrence of new targets. The SMC is used to track existing targets over time. The Mean-Shift clustering is adopted to automatically identify new comers. With the Monte Carlo based framework, the detection and tracking of multiple targets can be achieved in a unified and seamless manner. The detection and tracking accuracy is evaluated by both synthesized videos and real videos. The experimental results show that the proposed system can successfully track a varying number of people accurately. 1

Abstract not found