Learning enhanced 3D models for vehicle tracking

This paper presents an enhanced hypothesis verification strategy for 3D object recognition. A new learning methodology is presented which integrates the traditional dichotomic object-centred and appearance-based representations in computer vision giving improved hypothesis verification under iconic matching. The "appearance" of a 3D object is learnt using an eigenspace representation obtained as it is tracked through a scene. The feature representation implicitly models the background and the objects observed enabling the segmentation of the objects from the background. The method is shown to enhance model-based tracking, particularly in the presence of clutter and occlusion, and to provide a basis for identification. The unified approach is discussed in the context of the traffic surveillance domain. The approach is demonstrated on real-world image sequences and compared to previous (edge-based) iconic evaluation techniques.

Visual surveillance for moving vehicles

An overview is given of a vision system for locating, recognising and tracking multiple vehicles, using an image sequence taken by a single camera mounted on a moving vehicle. The camera motion is estimated by matching features on the ground plane from one image to the next. Vehicle detection and hypothesis generation are performed using template correlation and a 3D wire frame model of the vehicle is fitted to the image. Once detected and identified, vehicles are tracked using dynamic filtering. A separate batch mode filter obtains the 3D trajectories of nearby vehicles over an extended time. Results are shown for a motorway image sequence.

Driving saccade to pursuit using image motion

Within the context of active vision, scant attention has been paid to the execution of motion saccades—rapid re-adjustments of the direction of gaze to attend to moving objects. In this paper we first develop a methodology for, and give real-time demonstrations of, the use of motion detection and segmentation processes to initiate capture saccades towards a moving object. The saccade is driven by both position and velocity of the moving target under the assumption of constant target velocity, using prediction to overcome the delay introduced by visual processing. We next demonstrate the use of a first order approximation to the segmented motion field to compute bounds on the time-to-contact in the presence of looming motion. If the bound falls below a safe limit, a panic saccade is fired, moving the camera away from the approaching object. We then describe the use of image motion to realize smooth pursuit, tracking using velocity information alone, where the camera is moved so as to null a single constant image motion fitted within a central image region. Finally, we glue together capture saccades with smooth pursuit, thus effecting changes in both what is being attended to and how it is being attended to. To couple the different visual activities of waiting, saccading, pursuing and panicking, we use a finite state machine which provides inherent robustness outside of visual processing and provides a means of making repeated exploration. We demonstrate in repeated trials that the transition from saccadic motion to tracking is more likely to succeed using position and velocity control, than when using position alone.

Application of a hybrid tracking algorithm to motion analysis

An algorithm for tracking multiple feature positions in a dynamic image sequence is presented. This is achieved using a combination of two trajectory-based methods, with the resulting hybrid algorithm exhibiting the advantages of both. An optimizing exchange algorithm is described which enables short feature paths to be tracked without prior knowledge of the motion being studied. The resulting partial trajectories are then used to initialize a fast predictor algorithm which is capable of rapidly tracking multiple feature paths. As this predictor algorithm becomes tuned to the feature positions being tracked, it is shown how the location of occluded or poorly detected features can be predicted. The results of applying this tracking algorithm to data obtained from real-world scenes are then presented.

Prediction of processed cheese instrumental texture and meltability by mid-infrared spectroscopy coupled with chemometric tools

The objective of this study was to determine the potential of mid-infrared spectroscopy coupled with multidimensional statistical analysis for the prediction of processed cheese instrumental texture and meltability attributes. Processed cheeses (n = 32) of varying composition were manufactured in a pilot plant. Following two and four weeks storage at 4 degrees C samples were analysed using texture profile analysis, two meltability tests (computer vision, Olson and Price) and mid-infrared spectroscopy (4000-640 cm(-1)). Partial least squares regression was used to develop predictive models for all measured attributes. Five attributes were successfully modelled with varying degrees of accuracy. The computer vision meltability model allowed for discrimination between high and low melt values (R-2 = 0.64). The hardness and springiness models gave approximate quantitative results (R-2 = 0.77) and the cohesiveness (R-2 = 0.81) and Olson and Price meltability (R-2 = 0.88) models gave good prediction results. (c) 2006 Elsevier Ltd. All rights reserved..

Benchmark datasets for detection and tracking

There is a rising demand for the quantitative performance evaluation of automated video surveillance. To advance research in this area, it is essential that comparisons in detection and tracking approaches may be drawn and improvements in existing methods can be measured. There are a number of challenges related to the proper evaluation of motion segmentation, tracking, event recognition, and other components of a video surveillance system that are unique to the video surveillance community. These include the volume of data that must be evaluated, the difficulty in obtaining ground truth data, the definition of appropriate metrics, and achieving meaningful comparison of diverse systems. This chapter provides descriptions of useful benchmark datasets and their availability to the computer vision community. It outlines some ground truth and evaluation techniques, and provides links to useful resources. It concludes by discussing the future direction for benchmark datasets and their associated processes.

Reactions to peripheral image motion using a head/eye platform

The authors demonstrate four real-time reactive responses to movement in everyday scenes using an active head/eye platform. They first describe the design and realization of a high-bandwidth four-degree-of-freedom head/eye platform and visual feedback loop for the exploration of motion processing within active vision. The vision system divides processing into two scales and two broad functions. At a coarse, quasi-peripheral scale, detection and segmentation of new motion occurs across the whole image, and at fine scale, tracking of already detected motion takes place within a foveal region. Several simple coarse scale motion sensors which run concurrently at 25 Hz with latencies around 100 ms are detailed. The use of these sensors are discussed to drive the following real-time responses: (1) head/eye saccades to moving regions of interest; (2) a panic response to looming motion; (3) an opto-kinetic response to continuous motion across the image and (4) smooth pursuit of a moving target using motion alone.

Smoothed histogram modification for image processing

The technique of constructing a transformation, or regrading, of a discrete data set such that the histogram of the transformed data matches a given reference histogram is commonly known as histogram modification. The technique is widely used for image enhancement and normalization. A method which has been previously derived for producing such a regrading is shown to be “best” in the sense that it minimizes the error between the cumulative histogram of the transformed data and that of the given reference function, over all single-valued, monotone, discrete transformations of the data. Techniques for smoothed regrading, which provide a means of balancing the error in matching a given reference histogram against the information lost with respect to a linear transformation are also examined. The smoothed regradings are shown to optimize certain cost functionals. Numerical algorithms for generating the smoothed regradings, which are simple and efficient to implement, are described, and practical applications to the processing of LANDSAT image data are discussed.

A survey of human motion analysis using depth imagery

Analysis of human behaviour through visual information has been a highly active research topic in the computer vision community. This was previously achieved via images from a conventional camera, but recently depth sensors have made a new type of data available. This survey starts by explaining the advantages of depth imagery, then describes the new sensors that are available to obtain it. In particular, the Microsoft Kinect has made high-resolution real-time depth cheaply available. The main published research on the use of depth imagery for analysing human activity is reviewed. Much of the existing work focuses on body part detection and pose estimation. A growing research area addresses the recognition of human actions. The publicly available datasets that include depth imagery are listed, as are the software libraries that can acquire it from a sensor. This survey concludes by summarising the current state of work on this topic, and pointing out promising future research directions.

The future of video analytics for surveillance and its ethical implications

The current state of the art and direction of research in computer vision aimed at automating the analysis of CCTV images is presented. This includes low level identification of objects within the field of view of cameras, following those objects over time and between cameras, and the interpretation of those objects’ appearance and movements with respect to models of behaviour (and therefore intentions inferred). The potential ethical problems (and some potential opportunities) such developments may pose if and when deployed in the real world are presented, and suggestions made as to the necessary new regulations which will be needed if such systems are not to further enhance the power of the surveillers against the surveilled.

Biologically-inspired robust motion segmentation using mutual information

This paper presents a neuroscience inspired information theoretic approach to motion segmentation. Robust motion segmentation represents a fundamental first stage in many surveillance tasks. As an alternative to widely adopted individual segmentation approaches, which are challenged in different ways by imagery exhibiting a wide range of environmental variation and irrelevant motion, this paper presents a new biologically-inspired approach which computes the multivariate mutual information between multiple complementary motion segmentation outputs. Performance evaluation across a range of datasets and against competing segmentation methods demonstrates robust performance.

Representing 3D shape and location

The 3D shape of an object and its 3D location have traditionally thought of as very separate entities, although both can be described within a single 3D coordinate frame. Here, 3D shape and location are considered as two aspects of a view-based approach to representing depth, avoiding the use of 3D coordinate frames.

Low rank representation on Riemannian manifold of symmetric positive definite matrices

Sparse coding aims to find a more compact representation based on a set of dictionary atoms. A well-known technique looking at 2D sparsity is the low rank representation (LRR). However, in many computer vision applications, data often originate from a manifold, which is equipped with some Riemannian geometry. In this case, the existing LRR becomes inappropriate for modeling and incorporating the intrinsic geometry of the manifold that is potentially important and critical to applications. In this paper, we generalize the LRR over the Euclidean space to the LRR model over a specific Rimannian manifold—the manifold of symmetric positive matrices (SPD). Experiments on several computer vision datasets showcase its noise robustness and superior performance on classification and segmentation compared with state-of-the-art approaches.