Broadcast audio (speech input) equipment for AM, FM and television, September, 1945 ...

Mode of access: Internet.

Probabilistic visual recognition of artificial landmarks for simultaneous localization and mapping

Probabilistic robotics most often applied to the problem of simultaneous localisation and mapping (SLAM), requires measures of uncertainty to accompany observations of the environment. This paper describes how uncertainty can be characterised for a vision system that locates coloured landmarks in a typical laboratory environment. The paper describes a model of the uncertainty in segmentation, the internal cameral model and the mounting of the camera on the robot. It explains the implementation of the system on a laboratory robot, and provides experimental results that show the coherence of the uncertainty model.

Representation and learning of visual information for pose recognition

Recovering position from sensor information is an important problem in mobile robotics, known as localisation. Localisation requires a map or some other description of the environment to provide the robot with a context to interpret sensor data. The mobile robot system under discussion is using an artificial neural representation of position. Building a geometrical map of the environment with a single camera and artificial neural networks is difficult. Instead it would be simpler to learn position as a function of the visual input. Usually when learning images, an intermediate representation is employed. An appropriate starting point for biologically plausible image representation is the complex cells of the visual cortex, which have invariance properties that appear useful for localisation. The effectiveness for localisation of two different complex cell models are evaluated. Finally the ability of a simple neural network with single shot learning to recognise these representations and localise a robot is examined.

Category learning induces position invariance of pattern recognition across the visual field

Human object recognition is considered to be largely invariant to translation across the visual field. However, the origin of this invariance to positional changes has remained elusive, since numerous studies found that the ability to discriminate between visual patterns develops in a largely location-specific manner, with only a limited transfer to novel visual field positions. In order to reconcile these contradicting observations, we traced the acquisition of categories of unfamiliar grey-level patterns within an interleaved learning and testing paradigm that involved either the same or different retinal locations. Our results show that position invariance is an emergent property of category learning. Pattern categories acquired over several hours at a fixed location in either the peripheral or central visual field gradually become accessible at new locations without any position-specific feedback. Furthermore, categories of novel patterns presented in the left hemifield are distinctly faster learnt and better generalized to other locations than those learnt in the right hemifield. Our results suggest that during learning initially position-specific representations of categories based on spatial pattern structure become encoded in a relational, position-invariant format. Such representational shifts may provide a generic mechanism to achieve perceptual invariance in object recognition.

Attentional shifting and the role of the dorsal pathway in visual word recognition

A substantial amount of evidence has been collected to propose an exclusive role for the dorsal visual pathway in the control of guided visual search mechanisms, specifically in the preattentive direction of spatial selection [Vidyasagar, T. R. (1999). A neuronal model of attentional spotlight: Parietal guiding the temporal. Brain Research and Reviews, 30, 66-76; Vidyasagar, T. R. (2001). From attentional gating in macaque primary visual cortex to dyslexia in humans. Progress in Brain Research, 134, 297-312]. Moreover, it has been suggested recently that the dorsal visual pathway is specifically involved in the spatial selection and sequencing required for orthographic processing in visual word recognition. In this experiment we manipulate the demands for spatial processing in a word recognition, lexical decision task by presenting target words in a normal spatial configuration, or where the constituent letters of each word are spatially shifted relative to each other. Accurate word recognition in the Shifted-words condition should demand higher spatial encoding requirements, thereby making greater demands on the dorsal visual stream. Magnetoencephalographic (MEG) neuroimaging revealed a high frequency (35-40 Hz) right posterior parietal activation consistent with dorsal stream involvement occurring between 100 and 300 ms post-stimulus onset, and then again at 200-400 ms. Moreover, this signal was stronger in the shifted word condition, compared to the normal word condition. This result provides neurophysiological evidence that the dorsal visual stream may play an important role in visual word recognition and reading. These results further provide a plausible link between early stage theories of reading, and the magnocellular-deficit theory of dyslexia, which characterises many types of reading difficulty. © 2006 Elsevier Ltd. All rights reserved.

Visual word recognition:The first half second

We used magnetoencephalography (MEG) to map the spatiotemporal evolution of cortical activity for visual word recognition. We show that for five-letter words, activity in the left hemisphere (LH) fusiform gyrus expands systematically in both the posterior-anterior and medial-lateral directions over the course of the first 500 ms after stimulus presentation. Contrary to what would be expected from cognitive models and hemodynamic studies, the component of this activity that spatially coincides with the visual word form area (VWFA) is not active until around 200 ms post-stimulus, and critically, this activity is preceded by and co-active with activity in parts of the inferior frontal gyrus (IFG, BA44/6). The spread of activity in the VWFA for words does not appear in isolation but is co-active in parallel with spread of activity in anterior middle temporal gyrus (aMTG, BA 21 and 38), posterior middle temporal gyrus (pMTG, BA37/39), and IFG. © 2004 Elsevier Inc. All rights reserved.

False recognition of objects in visual scenes:findings from a combined direct and indirect memory test

We report an extension of the procedure devised by Weinstein and Shanks (Memory & Cognition 36:1415-1428, 2008) to study false recognition and priming of pictures. Participants viewed scenes with multiple embedded objects (seen items), then studied the names of these objects and the names of other objects (read items). Finally, participants completed a combined direct (recognition) and indirect (identification) memory test that included seen items, read items, and new items. In the direct test, participants recognized pictures of seen and read items more often than new pictures. In the indirect test, participants' speed at identifying those same pictures was improved for pictures that they had actually studied, and also for falsely recognized pictures whose names they had read. These data provide new evidence that a false-memory induction procedure can elicit memory-like representations that are difficult to distinguish from "true" memories of studied pictures. © 2012 Psychonomic Society, Inc.

How different are the visual representations used for object recognition in middle childhood and adulthood?

Recent experimental studies have shown that development towards adult performance levels in configural processing in object recognition is delayed through middle childhood. Whilst partchanges to animal and artefact stimuli are processed with similar to adult levels of accuracy from 7 years of age, relative size changes to stimuli result in a significant decrease in relative performance for participants aged between 7 and 10. Two sets of computational experiments were run using the JIM3 artificial neural network with adult and 'immature' versions to simulate these results. One set progressively decreased the number of neurons involved in the representation of view-independent metric relations within multi-geon objects. A second set of computational experiments involved decreasing the number of neurons that represent view-dependent (nonrelational) object attributes in JIM3's Surface Map. The simulation results which show the best qualitative match to empirical data occurred when artificial neurons representing metric-precision relations were entirely eliminated. These results therefore provide further evidence for the late development of relational processing in object recognition and suggest that children in middle childhood may recognise objects without forming structural description representations.

Accent Recognition for Noisy Audio Signals

It is well established that accent recognition can be as accurate as up to 95% when the signals are noise-free, using feature extraction techniques such as mel-frequency cepstral coefficients and binary classifiers such as discriminant analysis, support vector machine and k-nearest neighbors. In this paper, we demonstrate that the predictive performance can be reduced by as much as 15% when the signals are noisy. Specifically, in this paper we perturb the signals with different levels of white noise, and as the noise become stronger, the out-of-sample predictive performance deteriorates from 95% to 80%, although the in-sample prediction gives overly-optimistic results. ACM Computing Classification System (1998): C.3, C.5.1, H.1.2, H.2.4., G.3.

Development of face recognition: Infancy to early childhood

Perception and recognition of faces are fundamental cognitive abilities that form a basis for our social interactions. Research has investigated face perception using a variety of methodologies across the lifespan. Habituation, novelty preference, and visual paired comparison paradigms are typically used to investigate face perception in young infants. Storybook recognition tasks and eyewitness lineup paradigms are generally used to investigate face perception in young children. These methodologies have introduced systematic differences including the use of linguistic information for children but not infants, greater memory load for children than infants, and longer exposure times to faces for infants than for older children, making comparisons across age difficult. Thus, research investigating infant and child perception of faces using common methods, measures, and stimuli is needed to better understand how face perception develops. According to predictions of the Intersensory Redundancy Hypothesis (IRH; Bahrick & Lickliter, 2000, 2002), in early development, perception of faces is enhanced in unimodal visual (i.e., silent dynamic face) rather than bimodal audiovisual (i.e., dynamic face with synchronous speech) stimulation. The current study investigated the development of face recognition across children of three ages: 5 – 6 months, 18 – 24 months, and 3.5 – 4 years, using the novelty preference paradigm and the same stimuli for all age groups. It also assessed the role of modality (unimodal visual versus bimodal audiovisual) and memory load (low versus high) on face recognition. It was hypothesized that face recognition would improve across age and would be enhanced in unimodal visual stimulation with a low memory load. Results demonstrated a developmental trend (F(2, 90) = 5.00, p = 0.009) with older children showing significantly better recognition of faces than younger children. In contrast to predictions, no differences were found as a function of modality of presentation (bimodal audiovisual versus unimodal visual) or memory load (low versus high). This study was the first to demonstrate a developmental improvement in face recognition from infancy through childhood using common methods, measures and stimuli consistent across age.

Using sonic enhancement to augment non-visual tabular navigation

More information is now readily available to computer users than at any time in human history; however, much of this information is often inaccessible to people with blindness or low-vision, for whom information must be presented non-visually. Currently, screen readers are able to verbalize on-screen text using text-to-speech (TTS) synthesis; however, much of this vocalization is inadequate for browsing the Internet. An auditory interface that incorporates auditory-spatial orientation was created and tested. For information that can be structured as a two-dimensional table, links can be semantically grouped as cells in a row within an auditory table, which provides a consistent structure for auditory navigation. An auditory display prototype was tested.^ Sixteen legally blind subjects participated in this research study. Results demonstrated that stereo panning was an effective technique for audio-spatially orienting non-visual navigation in a five-row, six-column HTML table as compared to a centered, stationary synthesized voice. These results were based on measuring the time- to-target (TTT), or the amount of time elapsed from the first prompting to the selection of each tabular link. Preliminary analysis of the TTT values recorded during the experiment showed that the populations did not conform to the ANOVA requirements of normality and equality of variances. Therefore, the data were transformed using the natural logarithm. The repeated-measures two-factor ANOVA results show that the logarithmically-transformed TTTs were significantly affected by the tonal variation method, F(1,15) = 6.194, p= 0.025. Similarly, the results show that the logarithmically transformed TTTs were marginally affected by the stereo spatialization method, F(1,15) = 4.240, p=0.057. The results show that the logarithmically transformed TTTs were not significantly affected by the interaction of both methods, F(1,15) = 1.381, p=0.258. These results suggest that some confusion may be caused in the subject when employing both of these methods simultaneously. The significant effect of tonal variation indicates that the effect is actually increasing the average TTT. In other words, the presence of preceding tones increases task completion time on average. The marginally-significant effect of stereo spatialization decreases the average log(TTT) from 2.405 to 2.264.^

Datasets, features, learning, and models in visual recognition

Visual recognition is a fundamental research topic in computer vision. This dissertation explores datasets, features, learning, and models used for visual recognition. In order to train visual models and evaluate different recognition algorithms, this dissertation develops an approach to collect object image datasets on web pages using an analysis of text around the image and of image appearance. This method exploits established online knowledge resources (Wikipedia pages for text; Flickr and Caltech data sets for images). The resources provide rich text and object appearance information. This dissertation describes results on two datasets. The first is Berg’s collection of 10 animal categories; on this dataset, we significantly outperform previous approaches. On an additional set of 5 categories, experimental results show the effectiveness of the method. Images are represented as features for visual recognition. This dissertation introduces a text-based image feature and demonstrates that it consistently improves performance on hard object classification problems. The feature is built using an auxiliary dataset of images annotated with tags, downloaded from the Internet. Image tags are noisy. The method obtains the text features of an unannotated image from the tags of its k-nearest neighbors in this auxiliary collection. A visual classifier presented with an object viewed under novel circumstances (say, a new viewing direction) must rely on its visual examples. This text feature may not change, because the auxiliary dataset likely contains a similar picture. While the tags associated with images are noisy, they are more stable when appearance changes. The performance of this feature is tested using PASCAL VOC 2006 and 2007 datasets. This feature performs well; it consistently improves the performance of visual object classifiers, and is particularly effective when the training dataset is small. With more and more collected training data, computational cost becomes a bottleneck, especially when training sophisticated classifiers such as kernelized SVM. This dissertation proposes a fast training algorithm called Stochastic Intersection Kernel Machine (SIKMA). This proposed training method will be useful for many vision problems, as it can produce a kernel classifier that is more accurate than a linear classifier, and can be trained on tens of thousands of examples in two minutes. It processes training examples one by one in a sequence, so memory cost is no longer the bottleneck to process large scale datasets. This dissertation applies this approach to train classifiers of Flickr groups with many group training examples. The resulting Flickr group prediction scores can be used to measure image similarity between two images. Experimental results on the Corel dataset and a PASCAL VOC dataset show the learned Flickr features perform better on image matching, retrieval, and classification than conventional visual features. Visual models are usually trained to best separate positive and negative training examples. However, when recognizing a large number of object categories, there may not be enough training examples for most objects, due to the intrinsic long-tailed distribution of objects in the real world. This dissertation proposes an approach to use comparative object similarity. The key insight is that, given a set of object categories which are similar and a set of categories which are dissimilar, a good object model should respond more strongly to examples from similar categories than to examples from dissimilar categories. This dissertation develops a regularized kernel machine algorithm to use this category dependent similarity regularization. Experiments on hundreds of categories show that our method can make significant improvement for categories with few or even no positive examples.

Cognitive and Visual Speech Contributions to Speech Perception in Challenging Listening Conditions

Speech perception routinely takes place in noisy or degraded listening environments, leading to ambiguity in the identity of the speech token. Here, I present one review paper and two experimental papers that highlight cognitive and visual speech contributions to the listening process, particularly in challenging listening environments. First, I survey the literature linking audiometric age-related hearing loss and cognitive decline and review the four proposed causal mechanisms underlying this link. I argue that future research in this area requires greater consideration of the functional overlap between hearing and cognition. I also present an alternative framework for understanding causal relationships between age-related declines in hearing and cognition, with emphasis on the interconnected nature of hearing and cognition and likely contributions from multiple causal mechanisms. I also provide a number of testable hypotheses to examine how impairments in one domain may affect the other. In my first experimental study, I examine the direct contribution of working memory (through a cognitive training manipulation) on speech in noise comprehension in older adults. My results challenge the efficacy of cognitive training more generally, and also provide support for the contribution of sentence context in reducing working memory load. My findings also challenge the ubiquitous use of the Reading Span test as a pure test of working memory. In a second experimental (fMRI) study, I examine the role of attention in audiovisual speech integration, particularly when the acoustic signal is degraded. I demonstrate that attentional processes support audiovisual speech integration in the middle and superior temporal gyri, as well as the fusiform gyrus. My results also suggest that the superior temporal sulcus is sensitive to intelligibility enhancement, regardless of how this benefit is obtained (i.e., whether it is obtained through visual speech information or speech clarity). In addition, I also demonstrate that both the cingulo-opercular network and motor speech areas are recruited in difficult listening conditions. Taken together, these findings augment our understanding of cognitive contributions to the listening process and demonstrate that memory, working memory, and executive control networks may flexibly be recruited in order to meet listening demands in challenging environments.