Visual Recognition Performance Research Articles

Recognizing Dynamic Scenes: Influence of Processing Orientation

From face recognition studies, it is known that instructions are able to change processing orientation of stimuli, leading to an impairment of recognition performance. The present study examined instructional influences on the visual recognition of dynamic scenes. A global processing orientation without any instruction was assumed to lead to highest recognition performance, whereas instructions focusing participants' attention on certain characteristics of the event should lead to a local processing orientation with an impairment of visual recognition performance as a direct consequence. Since the pattern of results provided evidence for this hypothesis, theoretical contributions were discussed.

Using Remote Vision: The Effects of Video Image Frame Rate on Visual Object Recognition Performance

The process of using remote vision was simulated in order to determine the effects of video image frame rate on the performance in visual recognition of stationary environmental hazards in the dynamic video footage of the pedestrian travel environment. The recognition performance was assessed against two different video image frame rate variations: 25 and 2 fps. The assessment included a range of objective and subjective criteria. The obtained results show that the effects of the frame rate variations on the performance are statistically insignificant. This paper belongs to the process of development of a novel system for navigation of visually impaired pedestrians. The navigation system includes a remote vision facility, and the visual recognition of the environmental hazards by the sighted human guide is a basic activity in aiding the visually impaired user of the system in mobility.

시각 음성인식을 위한 영상 기반 접근방법에 기반한 강인한 시각 특징 파라미터의 추출 방법

In spite of development in speech recognition technology, speech recognition under noisy environment is still a difficult task. To solve this problem, Researchers has been proposed different methods where they have been used visual information except audio information for visual speech recognition. However, visual information also has visual noises as well as the noises of audio information, and this visual noises cause degradation in visual speech recognition. Therefore, it is one the field of interest how to extract visual features parameter for enhancing visual speech recognition performance. In this paper, we propose a method for visual feature parameter extraction based on image-base approach for enhancing recognition performance of the HMM based visual speech recognizer. For experiments, we have constructed Audio-visual database which is consisted with 105 speackers and each speaker has uttered 62 words. We have applied histogram matching, lip folding, RASTA filtering, Liner Mask, DCT and PCA. The experimental results show that the recognition performance of our proposed method enhanced at about 21% than the baseline method.

Item performance in visual word recognition

Standard factorial designs in psycholinguistics have been complemented recently by large-scale databases providing empirical constraints at the level of item performance. At the same time, the development of precise computational architectures has led modelers to compare item-level performance with item-level predictions. It has been suggested, however, that item performance includes a large amount of undesirable error variance that should be quantified to determine the amount of reproducible variance that models should account for. In the present study, we provide a simple and tractable statistical analysis of this issue. We also report practical solutions for estimating the amount of reproducible variance for any database that conforms to the additive decomposition of the variance. A new empirical database consisting of the word identification times of 140 participants on 120 words is then used to test these practical solutions. Finally, we show that increases in the amount of reproducible variance are accompanied by the detection of new sources of variance.

Modulation of brain activity by multiple lexical and word form variables in visual word recognition: A parametric fMRI study

Psycholinguistic research has documented a range of variables that influence visual word recognition performance. Many of these variables are highly intercorrelated. Most previous studies have used factorial designs, which do not exploit the full range of values available for continuous variables, and are prone to skewed stimulus selection as well as to effects of the baseline (e.g. when contrasting words with pseudowords). In our study, we used a parametric approach to study the effects of several psycholinguistic variables on brain activation. We focussed on the variable word frequency, which has been used in numerous previous behavioural, electrophysiological and neuroimaging studies, in order to investigate the neuronal network underlying visual word processing. Furthermore, we investigated the variable orthographic typicality as well as a combined variable for word length and orthographic neighbourhood size (N), for which neuroimaging results are still either scarce or inconsistent. Data were analysed using multiple linear regression analysis of event-related fMRI data acquired from 21 subjects in a silent reading paradigm. The frequency variable correlated negatively with activation in left fusiform gyrus, bilateral inferior frontal gyri and bilateral insulae, indicating that word frequency can affect multiple aspects of word processing. N correlated positively with brain activity in left and right middle temporal gyri as well as right inferior frontal gyrus. Thus, our analysis revealed multiple distinct brain areas involved in visual word processing within one data set.

Auditory Speech Recognition and Visual Text Recognition in Younger and Older Adults: Similarities and Differences Between Modalities and the Effects of Presentation Rate

To examine age-related differences in auditory speech recognition and visual text recognition performance for parallel sets of stimulus materials in the auditory and visual modalities. In addition, the effects of variation in rate of presentation of stimuli in each modality were investigated in each age group. A mixed-model design was used in which 3 independent groups (13 young adults with normal hearing, 10 elderly adults with normal hearing, and 16 elderly hearing-impaired adults) listened to auditory speech tests (a sentence-in-noise task, time-compressed monosyllables, and a speeded-spelling task) and viewed visual text-based analogs of the auditory tests. All auditory speech materials were presented so that the amplitude of the speech signal was at least 15 dB above threshold through 4000 Hz. Analyses of the group data revealed that when baseline levels of performance were used as covariates in the group analyses the only significant group difference was that both elderly groups performed worse than the young group on the auditory speeded-speech tasks. Analysis of individual data, using correlations, factor analysis, and linear regression, was generally consistent with the group data and revealed significant, moderate correlations of performance for similar tasks across modalities, but stronger correlations across tasks within a modality. This suggests that performance on these tasks was mediated both by a common underlying factor, such as cognitive processing, as well as modality-specific processing. Performance on measures of auditory processing of speech examined here was closely associated with performance on parallel measures of the visual processing of text obtained from the same participants. Young and older adults demonstrated comparable abilities in the use of contextual information in each modality, but older adults, regardless of hearing status, had more difficulty with fast presentation of auditory speech stimuli than young adults. There were no differences among the 3 groups with regard to the effects of presentation rate for the visual recognition of text, at least for the rates of presentation used here.

Nonvisual Motor Training Influences Biological Motion Perception

Experimental evidence suggests a link between perception and the execution of actions . In particular, it has been proposed that motor programs might directly influence visual action perception . According to this hypothesis, the acquisition of novel motor behaviors should improve their visual recognition, even in the absence of visual learning. We tested this prediction by using a new experimental paradigm that dissociates visual and motor learning during the acquisition of novel motor patterns. The visual recognition of gait patterns from point-light stimuli was assessed before and after nonvisual motor training. During this training, subjects were blindfolded and learned a novel coordinated upper-body movement based only on verbal and haptic feedback. The learned movement matched one of the visual test patterns. Despite the absence of visual stimulation during training, we observed a selective improvement of the visual recognition performance for the learned movement. Furthermore, visual recognition performance after training correlated strongly with the accuracy of the execution of the learned motor pattern. These results prove, for the first time, that motor learning has a direct and highly selective influence on visual action recognition that is not mediated by visual learning.

Visual sharpness contingency in recognition memory for orientation: Mnemonic illusion suppressed by sensory signature.

A striking finding about human memory is that people's level of accuracy in remembering the orientation of heads on coins is often not simply at the chance level but significantly below it. However, S. W. Kelly, A. M. Burton, T. Kato, and S. Akamatsu (2001) reported that this is not so when two-alternative forced-choice visual recognition is employed. The Kelly et al. result could not be replicated here with a copy of their stimuli. However, the result was successfully replicated with newly created stimuli. A series of experiments provided converging evidence that the mnemonic illusion is suppressed when recognition alternatives possess sharp visual detail. The role of a sensory signature in suppressing the mnemonic illusion and in modulating visual recognition performance in general is delineated.

Effects of cholinergic deafferentation of the rhinal cortex on visual recognition memory in monkeys

Excitotoxic lesion studies have confirmed that the rhinal cortex is essential for visual recognition ability in monkeys. To evaluate the mnemonic role of cholinergic inputs to this cortical region, we compared the visual recognition performance of monkeys given rhinal cortex infusions of a selective cholinergic immunotoxin, ME20.4-SAP, with the performance of monkeys given control infusions into this same tissue. The immunotoxin, which leads to selective cholinergic deafferentation of the infused cortex, yielded recognition deficits of the same magnitude as those produced by excitotoxic lesions of this region, providing the most direct demonstration to date that cholinergic activation of the rhinal cortex is essential for storing the representations of new visual stimuli and thereby enabling their later recognition.

Recent advances in visual and infrared face recognition—a review

Face recognition is a rapidly growing research area due to increasing demands for security in commercial and law enforcement applications. This paper provides an up-to-date review of research efforts in face recognition techniques based on two-dimensional (2D) images in the visual and infrared (IR) spectra. Face recognition systems based on visual images have reached a significant level of maturity with some practical success. However, the performance of visual face recognition may degrade under poor illumination conditions or for subjects of various skin colors. IR imagery represents a viable alternative to visible imaging in the search for a robust and practical identification system. While visual face recognition systems perform relatively reliably under controlled illumination conditions, thermal IR face recognition systems are advantageous when there is no control over illumination or for detecting disguised faces. Face recognition using 3D images is another active area of face recognition, which provides robust face recognition with changes in pose. Recent research has also demonstrated that the fusion of different imaging modalities and spectral components can improve the overall performance of face recognition.

Should visual speech cues (speechreading) be considered when fitting hearing aids?

When talker and listener are face-to-face, visual speech cues become an important part of the communication environment, and yet, these cues are seldom considered when designing hearing aids. Models of auditory–visual speech recognition highlight the importance of complementary versus redundant speech information for predicting auditory–visual recognition performance. Thus, for hearing aids to work optimally when visual speech cues are present, it is important to know whether the cues provided by amplification and the cues provided by speechreading complement each other. In this talk, data will be reviewed that show nonmonotonicity between auditory-alone speech recognition and auditory–visual speech recognition, suggesting that efforts designed solely to improve auditory-alone recognition may not always result in improved auditory–visual recognition. Data will also be presented showing that one of the most important speech cues for enhancing auditory–visual speech recognition performance, voicing, is often the cue that benefits least from amplification.

Developing Normal Reading Skills: Aspects of the Visual Processes Underlying Word Recognition

Visual word recognition performance of first graders (mean age: 6.6 years) through fifth graders (mean age: 10.8 years) was investigated using an experimental technique that is known to elicit the “viewing position effect” in skilled readers. The results showed that this effect, which consists of a systematic variation of performance as a function of fixation position within words, emerged early at the end of the 1st year of reading instruction. Visual field asymmetries in recognizing individual letters in words were also observed starting from first grade. Effects of word familiarity were obtained as early as in second grade. In contrast to skilled readers, children showed a marked word-length effect, which persisted through the first 5 years of instruction. No other qualitative differences between beginning and skilled readers were apparent. Hence, the basics of reading skills, as measured by the present technique, seem to be attained very early during acquisition. Further experience mainly reduces the time a reader needs to extract visual information from print.

Hypothetical reasoning and brainware

An enormous variety of complex problems requiring tentative interpretation of real-world situations or conditions, such as the visual recognition of natural scenes, as the basis for some action cannot be properly managed without the effective use of acquired knowledge to reduce the complexity dramatically. There is much evidence that the brain solves this type of problem by creating an internal hypothesis from its store of acquired knowledge. It then reformulates the task as essentially one of simply comparing this internally generated hypothesis with the objective sensory reality signaled to it by the sensors (such as the eye). Despite a rich body of research on hypothetical reasoning in the field of AI, this method of generating and verifying a hypothesis bridging signal and symbol levels has never been demonstrated. A neural control architecture that uses this problem-solving methodology is proposed in this paper. Multilayer neural network architecture, modeled after the essential features of the cortical structure and function, is used to simulate hypothetical reasoning and in particular to demonstrate its role in the marvelous performance of visual recognition. With this architecture it is possible not only to control the logical reasoning steps to ensure a rapid convergence in the decision-making processes but also to set the constraints for the self-organization of knowledge required in creating the initial internal hypothesis.

Synaptic depression induced by pharmacological activation of metabotropic glutamate receptors in the perirhinal cortex in vitro

The perirhinal cortex is crucially involved in various forms of learning and memory. Decrements in neuronal responsiveness occur in the perirhinal cortex with stimulus repetition during visual recognition performance. However, very little is known concerning the underlying mechanisms of synaptic transmission and plasticity in this cortical region. In this study, we provide evidence demonstrating the presence of functional group I, II and III metabotropic glutamate receptors in the rat perirhinal cortex in vitro. Furthermore, the results demonstrate long-lasting synaptic depression in the perirhinal cortex. Extracellular synaptic responses were recorded from superficial layers of the perirhinal cortex directly below the rhinal sulcus, in response to electrical stimuli delivered in the superficial or intermediate layers to the entorhinal or temporal cortex sides of the rhinal sulcus. Evoked synaptic potentials were depressed during bath perfusion of each of the following: the broad-spectrum metabotropic glutamate receptor agonist (1 S,3 R)-1-aminocyclopentane-1,3-dicarboxylic acid, the selective group I agonist ( R, S)-3,5-dihydroxyphenylglycine, the group II agonist (2 S,1′ R,2′ R,3′ R)-(2′,3′-dicarboxycyclopropyl)glycine and the group III agonist ( S)-2-amino-4-phosphonobutanoate. Furthermore, there was a long-lasting depression of synaptic transmission following washout of (1 S,3 R)-1-aminocyclopentane-1,3-dicarboxylic acid, ( R, S)-3,5-dihydroxyphenylglycine or (2 S,1′ R,2′ R,3′ R)-(2′,3′-dicarboxycyclopropyl)glycine. Activation of group III metabotropic glutamate receptors by ( S)-2-amino-4-phosphonobutanoate did not result in long-lasting changes in synaptic transmission. Thus, the pharmacological activation of metabotropic glutamate receptors can produce short- or long-term changes in synaptic transmission in the perirhinal cortex. It is possible therefore, that metabotropic glutamate receptors are involved in the decrement in neuronal responsiveness associated with visual recognition in the perirhinal cortex.

Sensitivity analysis of neocognitron

Fukushima's (1988; 1989; 1992; 1993) neocognitron model is well-known for its performance in visual pattern recognition. Through a training process, the visual pattern information is stored in a form of numerical weights in memory. When the model is actually implemented in hardware, weight errors and input noises caused by hardware imprecision and imperfect input devices respectively cannot be avoided and consequently the recognition performance usually degrades substantially from the theoretical result. In this paper, the effects of weight imprecision and input noise to the recognition performance of neocognitron are studied through a sensitivity analysis of the model. The sensitivity of an S-cell to weight and input perturbations is first derived, as a function of the weight and input perturbation ratios. An algorithm is proposed to combine the sensitivities of the S-cells in different layers to form the overall sensitivity of the model. The established sensitivity measure is then demonstrated to be a useful tool for hardware design. In addition, it has been found that the decision (recognition) error of neocognitron increases with the weight perturbation, the input perturbation, and the number of weights per neuron. This is similar to the result obtained for the Madaline model. Another important result is that decision error increases with the threshold/selectivity parameter. This supports the functional description of the threshold reported by Fukushina.

A computational theory of visual attention.

A computational theory of visual attention is presented. The basic theory (TVA) combines the biased-choice model for single-stimulus recognition with the fixed-capacity independent race model (FIRM) for selection from multi-element displays. TVA organizes a large body of experimental findings on performance in visual recognition and attention tasks. A recent development (CTVA) combines TVA with a theory of perceptual grouping by proximity. CTVA explains effects of perceptual grouping and spatial distance between items in multi-element displays. A new account of spatial focusing is proposed in this paper. The account provides a framework for understanding visual search as an interplay between serial and parallel processes.

Evaluation of convolutional neural networks for visual recognition

Convolutional neural networks provide an efficient method to constrain the complexity of feedforward neural networks by weight sharing and restriction to local connections. This network topology has been applied in particular to image classification when sophisticated preprocessing is to be avoided and raw images are to be classified directly. In this paper two variations of convolutional networks--neocognitron and a modification of neocognitron--are compared with classifiers based on fully connected feedforward layers (i.e., multilayer perceptron, nearest neighbor classifier, auto-encoding network) with respect to their visual recognition performance. Beside the original neocognitron a modification of the neocognitron is proposed which combines neurons from perceptron with the localized network structure of neocognitron. Instead of training convolutional networks by time-consuming error backpropagation, in this work a modular procedure is applied whereby layers are trained sequentially from the input to the output layer in order to recognize features of increasing complexity. For a quantitative experimental comparison with standard classifiers two very different recognition tasks have been chosen: handwritten digit recognition and face recognition. In the first example on handwritten digit recognition the generalization of convolutional networks is compared to fully connected networks. In several experiments the influence of variations of position, size, and orientation of digits is determined and the relation between training sample size and validation error is observed. In the second example recognition of human faces is investigated under constrained and variable conditions with respect to face orientation and illumination and the limitations of convolutional networks are discussed.

Visual Selective Attention: Outlines of a Choice Model, a Race Model and a Computational Theory

A computational theory of visual selective attention is presented. The theory developed out of a choice and a race model for visual selection from multi-elementdisplays. The choice model (Bundesen, Pedersen, & Larsen, 1984) provides a rule for computation of selection probabilities, which accounts for effects of the selection criterion and the numbers of targets and distractors in the stimulus. The race model (Shibuya & Bundesen, 1988) provides a process interpretation of the choice model and accounts for effects of the exposure duration of the stimulus. The computational theory (TVA; Bundesen, 1990) was constructed by integrating the race model with a biased-choice model for single-stimulus recognition (Luce, 1963). TVA describes two mechanisms (filtering and pigeonholing) by which selection is assumed to be carried out, and it organizes a large body of empirical data on human performance in visual recognition and attention tasks. A recent theoretical development (CTVA; Logan, 1996; Logan & Bundesen, 1996) combines TVA with a theory of perceptual grouping by proximity. CTVA explains effects of spatial separation between items in multi-element displays. The neural localization of the operations described in TVA is considered in the final section.

Occupation during life and memory performance in nondemented French elderly community residents.

Using data obtained during the baseline screening of the PAQUID cohort, an epidemiologic study on brain aging conducted in the Bordeaux area of France, we studied the relationship between lifetime occupation and memory performance in visual recognition (Benton Visual Retention Test; BVRT) and verbal induced recall (Wechsler Paired Associates Test; WPAT) in 2,720 elderly community residents. Occupational category was strongly correlated with poor memory performance for both the BVRT and the WPAT. Independent of the educational level, the risk of having poor memory performance was two to three times greater for farmers, domestic service employees, and blue-collar workers than for professionals/managerials. Occupational category should be taken into account in future cross-national studies on brain aging.

Similarity of Tactual and Visual Picture Recognition with Limited Field of View

Subjects attempted to recognize simple line drawings of common objects using either touch or vision. In the touch condition, subjects explored raised line drawings using the distal pad of the index finger or the distal pads both of the index and of the middle fingers. In the visual condition, a computer-driven display was used to simulate tactual exploration. By moving an electronic pen over a digitizing tablet, the subject could explore a line drawing stored in memory; on the display screen a portion of the drawing appeared to move behind a stationary aperture, in concert with the movement of the pen. This aperture was varied in width, thus simulating the use of one or two fingers. In terms of average recognition accuracy and average response latency, recognition performance was virtually the same in the one-finger touch condition and the simulated one-finger vision condition. Visual recognition performance improved considerably when the visual field size was doubled (simulating two fingers), but tactual performance showed little improvement, suggesting that the effective tactual field of view for this task is approximately equal to one finger pad. This latter result agrees with other reports in the literature indicating that integration of two-dimensional pattern information extending over multiple fingers on the same hand is quite poor. The near equivalence of tactual picture perception and narrow-field vision suggests that the difficulties of tactual picture recognition must be largely due to the narrowness of the effective field of view.