Search

The goal of object retrieval is to rank a set of images by the similarity of their contents to those of a query image. However, it is difficult to measure image content similarity due to visual changes caused by varying viewpoint and environment. In this paper, we propose a simple, efficient method to more effectively measure content similarity from image measurements. Our method is based on the ranking information available from existing retrieval systems. We observe that images within the set which, when used as queries, yield similar ranking lists are likely to be relevant to each other and vice versa. In our method, ranking consistency is used as a verification method to efficiently refine an existing ranking list, in much the same fashion that spatial verification is employed. The efficiency of our method is achieved by a list-wise min-Hash scheme, which allows rapid calculation of an approximate similarity ranking. Experimental results demonstrate the effectiveness of the proposed framework and its applications. HighlightsWe propose an image matching and retrieval framework based on ranking consistency.A list-wise min-Hash method ensures the efficiency of ranking verification.Our method is flexible and thus has a variety of retrieval-related applications.

A key enabler for the construction of large-scale intelligent surveillance systems is the accurate estimation of activity topology graphs. An activity topology graph describes the relationships between the fields of view of the cameras in a surveillance network. An accurate activity topology estimate allows higher-level processing such as network-wide tracking to be localised within neighbourhoods defined by the topology, and thus to scale. The camera overlap graph is an important special case of the general activity topology, in which edges represent overlap between cameras’ fields of view. We describe a family of pairwise occupancy overlap estimators, which are the only approaches proven to scale to networks with thousands of cameras. A distributed implementation is described, which enables the estimator to scale beyond the limits achievable by centralised implementations, and supports growth of the network whilst it remains online. Formulae are derived to describe the memory and network bandwidth requirements of the distributed implementation, which are verified by empirical results. Finally, the efficacy of the overlap estimators is demonstrated using results from their application in higher-level processing, specifically to network-wide tracking, which becomes feasible within the topology oriented architecture.

An accurate estimate of camera overlap is a key enabler for efficient network-wide surveillance processing (e.g. inter-camera tracking), especially in large-scale surveillance networks. Techniques based on contradictions in pair-wise occupancy data, such as the exclusion approach, have advantages in robustness and efficiency that make them particularly well suited for large surveillance networks. Correlation techniques share some of these advantages,but have a better understood statistical basis. This paper evaluates a set of contradiction and correlation techniques, using a novel metric, search space precision-recall. This metric reflects the activity-based overlap estimation required for camera handover, such as would be used in inter-camera tracking.Results are reported for a range of networks, including a 24-camera network setup in an office space, where the exclusion estimator showed the best performance.

The increased flexibility and other benefits offered by IP network cameras makes them a common choice for installation in new and expanded surveillance networks. One commonly quoted limitation of IP cameras is their high latency when compared to their analogue counterparts. This causes some reluctance to install or upgrade to digital cameras, and is slowing the adoption of live, intelligent analysis techniques in video surveillance systems. This paper presents methods for measurement of the latency in systems based upon digital IP or analogue cameras. These methods are camera-agnostic and require no specialised hardware. We use these methods to compare a variety of camera models. The results demonstrate that whilst analogue cameras do have a lower latency, most IP cameras are within acceptable tolerances. The source of the latency within an IP camera is also analysed, with prospects for improvement identified.

Large-scale intelligent video surveillance requires an accurate estimate of the relationships between the fields of view of the cameras in the network. The exclusion approach is the only method currently capable of performing online estimation of camera overlap for networks of more than 100 cameras, and implementations have demonstrated the capability to support networks of 1000 cameras. However, these implementations include a centralised processing component, with the practical result that the resources (in particular, memory) of the central processor limit the size of the network that can be supported. In this paper, we describe a new, partitioned, implementation of exclusion, suitable for deployment to a cluster of commodity servers. Results for this implementation demonstrate support for significantly larger camera networks than was previously feasible. Furthermore, the nature of the partitioning scheme enables incremental extension of system capacity through the addition of more servers, without interrupting the existing system. Finally, formulae for requirements of system memory and bandwidth resources, verified by experimental results, are derived to assist engineers seeking to implement the technique.

Making intelligent decisions on the basis of the video captured by a large network of surveillance cameras requires the ability to identify overlap between their fields of view. Without this information it is impossible to perform even simple analysis, such as distinguishing between repeated behaviours and multiple views of the same behaviour. Large-scale intelligent video surveillance thus requires a means of understanding the relationships between the fields of view of the cameras involved. The exclusion approach is the only method currently capable of performing online estimation of camera overlap for networks of more than 50 cameras, with a version of the algorithm applicable to 1000 camera networks having been published. Empirical evaluation of every such algorithm is critical to assessing its performance, and essential if comparisons between methods are to be made. This paper presents a method by which such an empirical evaluation may be carried out, and makes publicly available the data (including ground truth) on which it based in order that competing methods might be compared equally. Precision vs recall curves are reported for a series of experiments comparing the results of exclusion to ground truth. These results demonstrate the strengths and limitations of the exclusion-based estimation process, but show that the performance of the method exceeds the requirements of surveillance applications.

The development of assignments for undergraduate teaching typically requires a compromise between what is achievable by an average student and what engages the interest of a more advanced member of the class. Selecting a suitable compromise is particularly problematic for undergraduate artificial intelligence (AI) courses which typically attempt to cover a very broad range of topics, without delving too deeply into the details. Ideally, a single problem would be selected whose solution could be approached with more than one technique covered in the course, enabling students to carry out a comparative analysis of performance. Robot soccer simulation has provided an interesting platform for artificial intelligence research and is increasingly being used as a teaching apparatus. There are a number of limitations with existing simulation methodologies for this purpose. Current robot soccer simulators are aimed at research groups where accuracy is paramount and all facets of the real system must be emulated. However, many of the intricacies of a real robot soccer player are inappropriate for a teaching environment, as they detract from desired learning outcomes. Consequently, there is a need for a simulation that employs a simplified set of game rules and dynamics. This paper describes the design and implementation of such a framework and presents experiences gained from its use as a third year practical.