Contrast- and illumination-invariant object recognition from active sensation

It has been suggested that the deleterious effect of contrast reversal on visual recognition is unique to faces, not objects. Here we show from priming, supervised category learning, and generalization that there is no such thing as general invariance of recognition of non-face objects against contrast reversal and, likewise, changes in direction of illumination. However, when recognition varies with rendering conditions, invariance may be restored, and effects of continuous learning may be reduced, by providing prior object knowledge from active sensation. Our findings suggest that the degree of contrast invariance achieved reflects functional characteristics of object representations learned in a task-dependent fashion.

Behavioural and neuroimaging investigations of the relationship between visual attention, affordance and action

In this thesis the relationship between visual attention, affordance and action was investigated using a combination of neuroimaging and behavioural studies. Neuronal activity and movement construction were assessed when individuals passively viewed or produced action towards stimuli varying in their affordance and/or attentional attributes. The main findings were: (i) the passive perception of both object and abstract visual patterns was associated with decreased alpha and/or beta activity in sensori-motor cortex, occipito-temporal cortex and cerebellum. These are brain regions associated with the planning and production of visually guided action; (ii) for object patterns, decreased alpha and beta activity was also observed in regions of superior parietal and premotor cortex. These regions contain neurons argued to be essential for matching hand kinematics with manipulate objects; and (iii) in both control participants and a deafferented individual, studies of planned and unplanned pointing manoeuvres revealed that the attentional bias of a stimulus was critical for fast, efficient action production whereas the affordance bias was critical in determining end-point accuracy. Taken together, these findings demonstrate that affordance is not a necessary prerequisite for the potential of motor codes. Rather, affordance enables the construction of motor responses that reflect object functionality and/or manipulability. They further demonstrate that visual attention is associated with the potentiation of motor codes. Indeed, directed visual attention would appear critical for speeded responses. These findings provide new insights into the roles of directed visual attention and affordance upon action.

Effect of disability glare on visual performance

The object of the study was to investigate, establish and quantify the relationship between contrast sensitivity, intraocular light scatter and glare. The aim was to establish the effects on vision, in an effort to provide a more comprehensive understanding of the visual world of subjects prone to increased light scatter in the eye. Disability glare refers to the reduction in visual performance produced by a glare source. The reduction in visual performance can be explained by intraocular scattered light producing a veiling luminance which is superimposed upon the retinal image. This veiling luminance lowers contrast thus sensitivity to the stimulus declines. The effect of glare of luminance and colour contrast sensitivity for young and elderly subjects was examined. For both age groups, disability glare was greatest for the red-green stimulus and least for the blue-yellow. The precise effect of a glare source on colour discrimination depends upon the interaction between the chromaticity of the glare source and that of the stimulus. The effect of a long wavelength pass (red) and a short wavelength pass filter (blue) on disability glare was examined. Disability glare was not significantly different with the red and blue filters, even in the presence of wavelength dependent scatter. An equation was derived which allowed an intrinsic Light Scatter Factor (LSF) to be determined for any given glare angle (Paulsson and Sjöstrand, 1980). Corrections to the formula to account for factors such as pupil size changes are unnecessary. The results confirm the suitability of measuring the LSF using contrast threshold with and without glare, provided that appropriate methods are used. Using this formula an investigation into the amount of wavelength dependent scatter indicated that wavelength dependent scatter in normal young, elderly or cataractous eyes is of little or no significance. Finally, it seemed desirable to investigate the effect ultraviolet (UV) radiation has on intraocular light scatter and subsequently visual performance. Overall the results indicated that the presence or absence of UV radiation has relatively little effect on visual function for the young, elderly or cataract patient.

A penny for your thoughts! Patterns of fMRI activity reveal the content and the spatial topography of visual mental images

Visual mental imagery is a complex process that may be influenced by the content of mental images. Neuropsychological evidence from patients with hemineglect suggests that in the imagery domain environments and objects may be represented separately and may be selectively affected by brain lesions. In the present study, we used functional magnetic resonance imaging (fMRI) to assess the possibility of neural segregation among mental images depicting parts of an object, of an environment (imagined from a first-person perspective), and of a geographical map, using both a mass univariate and a multivariate approach. Data show that different brain areas are involved in different types of mental images. Imagining an environment relies mainly on regions known to be involved in navigational skills, such as the retrosplenial complex and parahippocampal gyrus, whereas imagining a geographical map mainly requires activation of the left angular gyrus, known to be involved in the representation of categorical relations. Imagining a familiar object mainly requires activation of parietal areas involved in visual space analysis in both the imagery and the perceptual domain. We also found that the pattern of activity in most of these areas specifically codes for the spatial arrangement of the parts of the mental image. Our results clearly demonstrate a functional neural segregation for different contents of mental images and suggest that visuospatial information is coded by different patterns of activity in brain areas involved in visual mental imagery. Hum Brain Mapp 36:945-958, 2015.

Visual signs and symptoms of dementia with Lewy bodies

Dementia with Lewy bodies ('Lewy body dementia' or 'diffuse Lewy body disease') (DLB) is the second most common form of dementia to affect elderly people, after Alzheimer's disease. A combination of the clinical symptoms of Alzheimer's disease and Parkinson's disease is present in DLB and the disorder is classified as a 'parkinsonian syndrome', a group of diseases which also includes Parkinson's disease, progressive supranuclear palsy, corticobasal degeneration and multiple system atrophy. Characteristics of DLB are fluctuating cognitive ability with pronounced variations in attention and alertness, recurrent visual hallucinations and spontaneous motor features, including akinesia, rigidity and tremor. In addition, DLB patients may exhibit visual signs and symptoms, including defects in eye movement, pupillary function and complex visual functions. Visual symptoms may aid the differential diagnoses of parkinsonian syndromes. Hence, the presence of visual hallucinations supports a diagnosis of Parkinson's disease or DLB rather than progressive supranuclear palsy. DLB and Parkinson's disease may exhibit similar impairments on a variety of saccadic and visual perception tasks (visual discrimination, space-motion and object-form recognition). Nevertheless, deficits in orientation, trail-making and reading the names of colours are often significantly greater in DLB than in Parkinson's disease. As primary eye-care practitioners, optometrists should be able to work with patients with DLB and their carers to manage their visual welfare.

Automatic object segmentation from calibrated images

This paper addresses the problem of automatically obtaining the object/background segmentation of a rigid 3D object observed in a set of images that have been calibrated for camera pose and intrinsics. Such segmentations can be used to obtain a shape representation of a potentially texture-less object by computing a visual hull. We propose an automatic approach where the object to be segmented is identified by the pose of the cameras instead of user input such as 2D bounding rectangles or brush-strokes. The key behind our method is a pairwise MRF framework that combines (a) foreground/background appearance models, (b) epipolar constraints and (c) weak stereo correspondence into a single segmentation cost function that can be efficiently solved by Graph-cuts. The segmentation thus obtained is further improved using silhouette coherency and then used to update the foreground/background appearance models which are fed into the next Graph-cut computation. These two steps are iterated until segmentation convergences. Our method can automatically provide a 3D surface representation even in texture-less scenes where MVS methods might fail. Furthermore, it confers improved performance in images where the object is not readily separable from the background in colour space, an area that previous segmentation approaches have found challenging. © 2011 IEEE.

Ingroup categorization affects the structural encoding of other-race faces:evidence from the N170 event-related potential

The current research examined the influence of ingroup/outgroup categorization on brain event-related potentials measured during perceptual processing of own- and other-race faces. White participants performed a sequential matching task with upright and inverted faces belonging either to their own race (White) or to another race (Black) and affiliated with either their own university or another university by a preceding visual prime. Results demonstrated that the right-lateralized N170 component evoked by test faces was modulated by race and by social category: the N170 to own-race faces showed a larger inversion effect (i.e., latency delay for inverted faces) when the faces were categorized as other-university rather than own-university members; the N170 to other-race faces showed no modulation of its inversion effect by university affiliation. These results suggest that neural correlates of structural face encoding (as evidenced by the N170 inversion effects) can be modulated by both visual (racial) and nonvisual (social) ingroup/outgroup status. © 2014 © 2014 Taylor & Francis.

Observation of a finger or an object movement primes imitative responses differentially

Behavioural advantages for imitation of human movements over movements instructed by other visual stimuli are attributed to an ‘action observation-execution matching’ (AOEM) mechanism. Here, we demonstrate that priming/exogenous cueing with a videotaped finger movement stimulus (S1) produces specific congruency effects in reaction times (RTs) of imitative responses to a target movement (S2) at defined stimulus onset asynchronies (SOAs). When contrasted with a moving object at an SOA of 533 ms, only a human movement is capable of inducing an effect reminiscent of ‘inhibition of return’ (IOR), i.e. a significant advantage for imitation of a subsequent incongruent as compared to a congruent movement. When responses are primed by a finger movement at SOAs of 533 and 1,200 ms, inhibition of congruent or facilitation of incongruent responses, respectively, is stronger as compared to priming by a moving object. This pattern does not depend on whether S2 presents a finger movement or a moving object, thus effects cannot be attributed to visual similarity between S1 and S2. We propose that, whereas both priming by a finger movement and a moving object induces processes of spatial orienting, solely observation of a human movement activates AOEM. Thus, S1 immediately elicits an imitative response tendency. As an overt imitation of S1 is inadequate in the present setting, the response is inhibited which, in turn, modulates congruency effects.

Teallach - a flexible user - interface development environment for object database applications

The Teallach project has adapted model-based user-interface development techniques to the systematic creation of user-interfaces for object-oriented database applications. Model-based approaches aim to provide designers with a more principled approach to user-interface development using a variety of underlying models, and tools which manipulate these models. Here we present the results of the Teallach project, describing the tools developed and the flexible design method supported. Distinctive features of the Teallach system include provision of database-specific constructs, comprehensive facilities for relating the different models, and support for a flexible design method in which models can be constructed and related by designers in different orders and in different ways, to suit their particular design rationales. The system then creates the desired user-interface as an independent, fully functional Java application, with automatically generated help facilities.

The effect of relative distance enlargement on visual acuity in the visually impaired

Background: Prescribing magnification is typically based on distance or near visual acuity. this presumes a constant minimum angle of visual resolution with working distance and therefore enlargement of an object moved to a shorter working distance (relative distance enlargement). this study examines this premise in a visually impaired population. methods: distance letter visual acuity was measured prospectively for 380 low vision patients (distance visual acuity between 0.3 and 2.1 logmar) over the age of 57 years, along with near word visual acuity at an appropriate distance for near lens additions from +4 d to +20 D. demographic information, the disease causing low vision, contrast sensitivity, visual field and psychological status were also recorded. results: distance letter acuity was significantly related to (r = 0.84) but on average 0.1 ± 0.2 logmar better (1 ± 2 lines on a logmar chart) than near word acuity at 25 cm with a +4 d lens addition. in 39. 8 per cent of patients, near word acuity was more than 0.1 logmar worse than distance letter acuity. in 11.0 per cent of subjects, near visual acuity was more than 0.1 logmar better than distance letter acuity. the group with near word acuity worse than distance letter acuity also had lower contrast sensitivity. the group with near word acuity better than distance letter acuity was less likely to have age-Related macular degeneration. smaller print size could be read by reducing working distance (achieved by using higher near lens additions) in 86. 1 per cent, although not by as much as predicted by geometric progression in 14. 5 per cent. discussion: although distance letter and near word acuity are highly related, they are on average 1 logmar line different and this varies significantly between individuals. near word acuity did not increase linearly with relative distance enlargement in approximately one in seven visually impaired, suggesting that the measurement of visual resolution over a range of working distances will assist appropriate prescribing of magnification aids.

Imagery in multi-modal object learning

Spatial objects may not only be perceived visually but also by touch. We report recent experiments investigating to what extent prior object knowledge acquired in either the haptic or visual sensory modality transfers to a subsequent visual learning task. Results indicate that even mental object representations learnt in one sensory modality may attain a multi-modal quality. These findings seem incompatible with picture-based reasoning schemas but leave open the possibility of modality-specific reasoning mechanisms.

Resolving ambiguous descriptions through visual information

The project “Reference in Discourse” deals with the selection of a specific object from a visual scene in a natural language situation. The goal of this research is to explain this everyday discourse reference task in terms of a concept generation process based on subconceptual visual and verbal information. The system OINC (Object Identification in Natural Communicators) aims at solving this problem in a psychologically adequate way. The system’s difficulties occurring with incomplete and deviant descriptions correspond to the data from experiments with human subjects. The results of these experiments are reported.

Incremental tensor biased discriminant analysis:a new color-based visual tracking method

Most existing color-based tracking algorithms utilize the statistical color information of the object as the tracking clues, without maintaining the spatial structure within a single chromatic image. Recently, the researches on the multilinear algebra provide the possibility to hold the spatial structural relationship in a representation of the image ensembles. In this paper, a third-order color tensor is constructed to represent the object to be tracked. Considering the influence of the environment changing on the tracking, the biased discriminant analysis (BDA) is extended to the tensor biased discriminant analysis (TBDA) for distinguishing the object from the background. At the same time, an incremental scheme for the TBDA is developed for the tensor biased discriminant subspace online learning, which can be used to adapt to the appearance variant of both the object and background. The experimental results show that the proposed method can track objects precisely undergoing large pose, scale and lighting changes, as well as partial occlusion. © 2009 Elsevier B.V.

Towards Active Vision in the DUAL Cognitive Architecture

The paper describes an extension of the cognitive architecture DUAL with a model of visual attention and perception. The goal of this attempt is to account for the construction and the categorization of object and scene representations derived from visual stimuli in the TextWorld microdomain. Low-level parallel computations are combined with an active serial deployment of visual attention enabling the construction of abstract symbolic representations. A limited-capacity short-term visual store holding information across attention shifts forms the core of the model interfacing between the low-level representation of the stimulus and DUAL’s semantic memory. The model is validated by comparing the results of a simulation with real data from an eye movement experiment with human subjects.

Self-organising zooms for decentralised redundancy management in visual sensor networks

When visual sensor networks are composed of cameras which can adjust the zoom factor of their own lens, one must determine the optimal zoom levels for the cameras, for a given task. This gives rise to an important trade-off between the overlap of the different cameras’ fields of view, providing redundancy, and image quality. In an object tracking task, having multiple cameras observe the same area allows for quicker recovery, when a camera fails. In contrast having narrow zooms allow for a higher pixel count on regions of interest, leading to increased tracking confidence. In this paper we propose an approach for the self-organisation of redundancy in a distributed visual sensor network, based on decentralised multi-objective online learning using only local information to approximate the global state. We explore the impact of different zoom levels on these trade-offs, when tasking omnidirectional cameras, having perfect 360-degree view, with keeping track of a varying number of moving objects. We further show how employing decentralised reinforcement learning enables zoom configurations to be achieved dynamically at runtime according to an operator’s preference for maximising either the proportion of objects tracked, confidence associated with tracking, or redundancy in expectation of camera failure. We show that explicitly taking account of the level of overlap, even based only on local knowledge, improves resilience when cameras fail. Our results illustrate the trade-off between maintaining high confidence and object coverage, and maintaining redundancy, in anticipation of future failure. Our approach provides a fully tunable decentralised method for the self-organisation of redundancy in a changing environment, according to an operator’s preferences.