Visual surveillance using deformable models of vehicles

This paper presents recent developments to a vision-based traffic surveillance system which relies extensively on the use of geometrical and scene context. Firstly, a highly parametrised 3-D model is reported, able to adopt the shape of a wide variety of different classes of vehicle (e.g. cars, vans, buses etc.), and its subsequent specialisation to a generic car class which accounts for commonly encountered types of car (including saloon, batchback and estate cars). Sample data collected from video images, by means of an interactive tool, have been subjected to principal component analysis (PCA) to define a deformable model having 6 degrees of freedom. Secondly, a new pose refinement technique using “active” models is described, able to recover both the pose of a rigid object, and the structure of a deformable model; an assessment of its performance is examined in comparison with previously reported “passive” model-based techniques in the context of traffic surveillance. The new method is more stable, and requires fewer iterations, especially when the number of free parameters increases, but shows somewhat poorer convergence. Typical applications for this work include robot surveillance and navigation tasks.

Interaction of visual and haptic information in simulated environments: texture perception

This paper describes experiments relating to the perception of the roughness of simulated surfaces via the haptic and visual senses. Subjects used a magnitude estimation technique to judge the roughness of “virtual gratings” presented via a PHANToM haptic interface device, and a standard visual display unit. It was shown that under haptic perception, subjects tended to perceive roughness as decreasing with increased grating period, though this relationship was not always statistically significant. Under visual exploration, the exact relationship between spatial period and perceived roughness was less well defined, though linear regressions provided a reliable approximation to individual subjects’ estimates.

View-based modelling of human visual navigation errors

View-based and Cartesian representations provide rival accounts of visual navigation in humans, and here we explore possible models for the view-based case. A visual “homing” experiment was undertaken by human participants in immersive virtual reality. The distributions of end-point errors on the ground plane differed significantly in shape and extent depending on visual landmark configuration and relative goal location. A model based on simple visual cues captures important characteristics of these distributions. Augmenting visual features to include 3D elements such as stereo and motion parallax result in a set of models that describe the data accurately, demonstrating the effectiveness of a view-based approach.

A demonstration of 'broken' visual space

It has long been assumed that there is a distorted mapping between real and ‘perceived’ space, based on demonstrations of systematic errors in judgements of slant, curvature, direction and separation. Here, we have applied a direct test to the notion of a coherent visual space. In an immersive virtual environment, participants judged the relative distance of two squares displayed in separate intervals. On some trials, the virtual scene expanded by a factor of four between intervals although, in line with recent results, participants did not report any noticeable change in the scene. We found that there was no consistent depth ordering of objects that can explain the distance matches participants made in this environment (e.g. A > B > D yet also A < C < D) and hence no single one-to-one mapping between participants’ perceived space and any real 3D environment. Instead, factors that affect pairwise comparisons of distances dictate participants’ performance. These data contradict, more directly than previous experiments, the idea that the visual system builds and uses a coherent 3D internal representation of a scene.

Towards a model-based integration of co-registered electroencephalography/functional magnetic resonance imaging data with realistic neural population meshes

Brain activity can be measured with several non-invasive neuroimaging modalities, but each modality has inherent limitations with respect to resolution, contrast and interpretability. It is hoped that multimodal integration will address these limitations by using the complementary features of already available data. However, purely statistical integration can prove problematic owing to the disparate signal sources. As an alternative, we propose here an advanced neural population model implemented on an anatomically sound cortical mesh with freely adjustable connectivity, which features proper signal expression through a realistic head model for the electroencephalogram (EEG), as well as a haemodynamic model for functional magnetic resonance imaging based on blood oxygen level dependent contrast (fMRI BOLD). It hence allows simultaneous and realistic predictions of EEG and fMRI BOLD from the same underlying model of neural activity. As proof of principle, we investigate here the influence on simulated brain activity of strengthening visual connectivity. In the future we plan to fit multimodal data with this neural population model. This promises novel, model-based insights into the brain's activity in sleep, rest and task conditions.

Modelling human visual navigation using multi-view scene reconstruction

It is often assumed that humans generate a 3D reconstruction of the environment, either in egocentric or world-based coordinates, but the steps involved are unknown. Here, we propose two reconstruction-based models, evaluated using data from two tasks in immersive virtual reality. We model the observer’s prediction of landmark location based on standard photogrammetric methods and then combine location predictions to compute likelihood maps of navigation behaviour. In one model, each scene point is treated independently in the reconstruction; in the other, the pertinent variable is the spatial relationship between pairs of points. Participants viewed a simple environment from one location, were transported (virtually) to another part of the scene and were asked to navigate back. Error distributions varied substantially with changes in scene layout; we compared these directly with the likelihood maps to quantify the success of the models. We also measured error distributions when participants manipulated the location of a landmark to match the preceding interval, providing a direct test of the landmark-location stage of the navigation models. Models such as this, which start with scenes and end with a probabilistic prediction of behaviour, are likely to be increasingly useful for understanding 3D vision.

Issues and benefits of using 3D Interfaces: visual and verbal tasks

Spatial memory is important for locating objects in hierarchical data structures, such as desktop folders. There are, however, some contradictions in literature concerning the effectiveness of 3D user interfaces when compared to their 2D counterparts. This paper uses a task-based approach in order to investigate the effectiveness of adding a third dimension to specific user tasks, i.e. the impact of depth on navigation in a 3D file manager. Results highlight issues and benefits of using 3D interfaces for visual and verbal tasks, and introduces the possible existence of a correlation between aptitude scores achieved on the Guilford- Zimmerman Orientation Survey and Electroencephalography- measured brainwave activity as participants search for targets of variable perceptual salience in 2D and 3D environments.

An exploration of the equivalent weights particle filter

Particle filters are fully non-linear data assimilation techniques that aim to represent the probability distribution of the model state given the observations (the posterior) by a number of particles. In high-dimensional geophysical applications the number of particles required by the sequential importance resampling (SIR) particle filter in order to capture the high probability region of the posterior, is too large to make them usable. However particle filters can be formulated using proposal densities, which gives greater freedom in how particles are sampled and allows for a much smaller number of particles. Here a particle filter is presented which uses the proposal density to ensure that all particles end up in the high probability region of the posterior probability density function. This gives rise to the possibility of non-linear data assimilation in large dimensional systems. The particle filter formulation is compared to the optimal proposal density particle filter and the implicit particle filter, both of which also utilise a proposal density. We show that when observations are available every time step, both schemes will be degenerate when the number of independent observations is large, unlike the new scheme. The sensitivity of the new scheme to its parameter values is explored theoretically and demonstrated using the Lorenz (1963) model.

Televised Whorf: cognitive restructuring in advanced foreign language learners as a function of audio-visual media exposure

The encoding of goal-oriented motion events varies across different languages. Speakers of languages without grammatical aspect (e.g., Swedish) tend to mention motion endpoints when describing events, e.g., “two nuns walk to a house,”, and attach importance to event endpoints when matching scenes from memory. Speakers of aspect languages (e.g., English), on the other hand, are more prone to direct attention to the ongoingness of motion events, which is reflected both in their event descriptions, e.g., “two nuns are walking.”, and in their non-verbal similarity judgements. This study examines to what extent native speakers of Swedish (n = 82) with English as a foreign language (FL) restructure their categorisation of goal-oriented motion as a function of their English proficiency and experience with the English language (e.g., exposure, learning). Seventeen monolingual native English speakers from the United Kingdom (UK) were engaged for comparison purposes. Data on motion event cognition were collected through a memory-based triads matching task, in which a target scene with an intermediate degree of endpoint orientation was matched with two alternative scenes with low and high degrees of endpoint orientation, respectively. Results showed that the preference among the Swedish speakers of L2 English to base their similarity judgements on ongoingness rather than event endpoints was correlated with their use of English in their everyday lives, such that those who often watched television in English approximated the ongoingness preference of the English native speakers. These findings suggest that event cognition patterns may be restructured through the exposure to FL audio-visual media. The results thus add to the emerging picture that learning a new language entails learning new ways of observing and reasoning about reality.

Visual sensor fusion for active security in robotic industrial environments

This work presents a method of information fusion involving data captured by both a standard CCD camera and a ToF camera to be used in the detection of the proximity between a manipulator robot and a human. Both cameras are assumed to be located above the work area of an industrial robot. The fusion of colour images and time of light information makes it possible to know the 3D localization of objects with respect to a world coordinate system. At the same time this allows to know their colour information. Considering that ToF information given by the range camera contains innacuracies including distance error, border error, and pixel saturation, some corrections over the ToF information are proposed and developed to improve the results. The proposed fusion method uses the calibration parameters of both cameras to reproject 3D ToF points, expressed in a common coordinate system for both cameras and a robot arm, in 2D colour images. In addition to this, using the 3D information, the motion detection in a robot industrial environment is achieved, and the fusion of information is applied to the foreground objects previously detected. This combination of information results in a matrix that links colour and 3D information, giving the possibility of characterising the object by its colour in addition to its 3D localization. Further development of these methods will make it possible to identify objects and their position in the real world, and to use this information to prevent possible collisions between the robot and such objects.

Visual sensor fusion for active security in robotic industrial environments

This work presents a method of information fusion involving data captured by both a standard charge-coupled device (CCD) camera and a time-of-flight (ToF) camera to be used in the detection of the proximity between a manipulator robot and a human. Both cameras are assumed to be located above the work area of an industrial robot. The fusion of colour images and time-of-flight information makes it possible to know the 3D localization of objects with respect to a world coordinate system. At the same time, this allows to know their colour information. Considering that ToF information given by the range camera contains innacuracies including distance error, border error, and pixel saturation, some corrections over the ToF information are proposed and developed to improve the results. The proposed fusion method uses the calibration parameters of both cameras to reproject 3D ToF points, expressed in a common coordinate system for both cameras and a robot arm, in 2D colour images. In addition to this, using the 3D information, the motion detection in a robot industrial environment is achieved, and the fusion of information is applied to the foreground objects previously detected. This combination of information results in a matrix that links colour and 3D information, giving the possibility of characterising the object by its colour in addition to its 3D localisation. Further development of these methods will make it possible to identify objects and their position in the real world and to use this information to prevent possible collisions between the robot and such objects.

Global burned-land estimation in Latin America using MODIS composite data

This paper presents results of the AQL2004 project, which has been develope within the GOFC-GOLD Latin American network of remote sensing and forest fires (RedLatif). The project intended to obtain monthly burned-land maps of the entire region, from Mexico to Patagonia, using MODIS (moderate-resolution imaging spectroradiometer) reflectance data. The project has been organized in three different phases: acquisition and preprocessing of satellite data; discrimination of burned pixels; and validation of results. In the first phase, input data consisting of 32-day composites of MODIS 500-m reflectance data generated by the Global Land Cover Facility (GLCF) of the University of Maryland (College Park, Maryland, U.S.A.) were collected and processed. The discrimination of burned areas was addressed in two steps: searching for "burned core" pixels using postfire spectral indices and multitemporal change detection and mapping of burned scars using contextual techniques. The validation phase was based on visual analysis of Landsat and CBERS (China-Brazil Earth Resources Satellite) images. Validation of the burned-land category showed an agreement ranging from 30% to 60%, depending on the ecosystem and vegetation species present. The total burned area for the entire year was estimated to be 153 215 km2. The most affected countries in relation to their territory were Cuba, Colombia, Bolivia, and Venezuela. Burned areas were found in most land covers; herbaceous vegetation (savannas and grasslands) presented the highest proportions of burned area, while perennial forest had the lowest proportions. The importance of croplands in the total burned area should be taken with reserve, since this cover presented the highest commission errors. The importance of generating systematic products of burned land areas for different ecological processes is emphasized.

Tests of sunspot number sequences: 3. Effects of regression procedures on the calibration of historic sunspot data

We use sunspot group observations from the Royal Greenwich Observatory (RGO) to investigate the effects of intercalibrating data from observers with different visual acuities. The tests are made by counting the number of groups RB above a variable cut-off threshold of observed total whole-spot area (uncorrected for foreshortening) to simulate what a lower acuity observer would have seen. The synthesised annual means of RB are then re-scaled to the full observed RGO group number RA using a variety of regression techniques. It is found that a very high correlation between RA and RB (rAB > 0.98) does not prevent large errors in the intercalibration (for example sunspot maximum values can be over 30 % too large even for such levels of rAB). In generating the backbone sunspot number (RBB), Svalgaard and Schatten (2015, this issue) force regression fits to pass through the scatter plot origin which generates unreliable fits (the residuals do not form a normal distribution) and causes sunspot cycle amplitudes to be exaggerated in the intercalibrated data. It is demonstrated that the use of Quantile-Quantile (“Q  Q”) plots to test for a normal distribution is a useful indicator of erroneous and misleading regression fits. Ordinary least squares linear fits, not forced to pass through the origin, are sometimes reliable (although the optimum method used is shown to be different when matching peak and average sunspot group numbers). However, other fits are only reliable if non-linear regression is used. From these results it is entirely possible that the inflation of solar cycle amplitudes in the backbone group sunspot number as one goes back in time, relative to related solar-terrestrial parameters, is entirely caused by the use of inappropriate and non-robust regression techniques to calibrate the sunspot data.

Are visual threats prioritized without awareness? A critical review and meta analysis involving 3 behavioral paradigms and 2696 observers

Given capacity limits, only a subset of stimuli 1 give rise to a conscious percept. Neurocognitive models suggest that humans have evolved mechanisms that operate without awareness and prioritize threatening stimuli over neutral stimuli in subsequent perception. In this meta analysis, we review evidence for this ‘standard hypothesis’ emanating from three widely used, but rather different experimental paradigms that have been used to manipulate awareness. We found a small pooled threat-bias effect in the masked visual probe paradigm, a medium effect in the binocular rivalry paradigm and highly inconsistent effects in the breaking continuous flash suppression paradigm. Substantial heterogeneity was explained by the stimulus type: the only threat stimuli that were robustly prioritized across all three paradigms were fearful faces. Meta regression revealed that anxiety may modulate threat biases, but only under specific presentation conditions. We also found that insufficiently rigorous awareness measures, inadequate control of response biases and low level confounds may undermine claims of genuine unconscious threat processing. Considering the data together, we suggest that uncritical acceptance of the standard hypothesis is premature: current behavioral evidence for threat-sensitive visual processing that operates without awareness is weak.

Environmental factors and features that influence visual search in a 3D WIMP interface

The challenge of moving past the classic Window Icons Menus Pointer (WIMP) interface, i.e. by turning it ‘3D’, has resulted in much research and development. To evaluate the impact of 3D on the ‘finding a target picture in a folder’ task, we built a 3D WIMP interface that allowed the systematic manipulation of visual depth, visual aides, semantic category distribution of targets versus non-targets; and the detailed measurement of lower-level stimuli features. Across two separate experiments, one large sample web-based experiment, to understand associations, and one controlled lab environment, using eye tracking to understand user focus, we investigated how visual depth, use of visual aides, use of semantic categories, and lower-level stimuli features (i.e. contrast, colour and luminance) impact how successfully participants are able to search for, and detect, the target image. Moreover in the lab-based experiment, we captured pupillometry measurements to allow consideration of the influence of increasing cognitive load as a result of either an increasing number of items on the screen, or due to the inclusion of visual depth. Our findings showed that increasing the visible layers of depth, and inclusion of converging lines, did not impact target detection times, errors, or failure rates. Low-level features, including colour, luminance, and number of edges, did correlate with differences in target detection times, errors, and failure rates. Our results also revealed that semantic sorting algorithms significantly decreased target detection times. Increased semantic contrasts between a target and its neighbours correlated with an increase in detection errors. Finally, pupillometric data did not provide evidence of any correlation between the number of visible layers of depth and pupil size, however, using structural equation modelling, we demonstrated that cognitive load does influence detection failure rates when there is luminance contrasts between the target and its surrounding neighbours. Results suggest that WIMP interaction designers should consider stimulus-driven factors, which were shown to influence the efficiency with which a target icon can be found in a 3D WIMP interface.