Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “vision transformers”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 163 records · Page 9Linked to original sources
PubMed2001

Estimation of an unexpected-overlooking error by means of the single eye fixation related potential analysis with wavelet transform filter.

An unexpected-overlooking error that caused failure to notice near the peripheral vision is one of the accident factors in driving behavior. We estimated how the unexpected-overlooking error affected the amplitude of the lambda wave in the eye fixation related potential (EFRP). Four subjects participated in the experiment. Each subject was required press the right or left switch according to the given task, which was that he/she pressed the right switch when the blue dot appeared in the right detected area or he/she pressed the left switch when the red dot appeared in the right. The single trial data from Pz, which referred to both earlobes, were analyzed by means of a wavelet transform (WT) filter. The difference of the lambda amplitude between the corrected data was applied for analysis of variance. Three subjects showed a significant effect (P<0.01 or P<0.05), and the remaining one subject did not show a significant consequence of only two errors. The unexpected-overlooking errors had a low amplitude compared to the mean of amplitude throughout the task. It was concluded that the amplitude of the lambda wave might reflect the attention level of a subject.

Algorithms↗
PubMed2006

Three-dimensional vision from a multisensing mechanism.

An approach for sensing a three-dimensional (3D) object surface with an arbitrary geometric shape is presented. Combining two different 3D sensing mechanisms, point-array encoding based on affine transformation and fringe encoding based on phase mapping, we construct a mathematic model for 3D vision in which the point-array encoding is initially applied to determine the fringe orders to create a control-vertex mesh with absolute coordinate values in 3D space. Then phase evaluation and phase unwrapping for fringe decoding is performed under the guidance of control-vertex mesh, leading to an absolute phase map and the corresponding range image of the test object. The computer simulations and experimental results are presented to demonstrate the theoretical prediction.

Journal Article↗
PubMed2005

Learning invariant object recognition in the visual system with continuous transformations.

The cerebral cortex utilizes spatiotemporal continuity in the world to help build invariant representations. In vision, these might be representations of objects. The temporal continuity typical of objects has been used in an associative learning rule with a short-term memory trace to help build invariant object representations. In this paper, we show that spatial continuity can also provide a basis for helping a system to self-organize invariant representations. We introduce a new learning paradigm "continuous transformation learning" which operates by mapping spatially similar input patterns to the same postsynaptic neurons in a competitive learning system. As the inputs move through the space of possible continuous transforms (e.g. translation, rotation, etc.), the active synapses are modified onto the set of postsynaptic neurons. Because other transforms of the same stimulus overlap with previously learned exemplars, a common set of postsynaptic neurons is activated by the new transforms, and learning of the new active inputs onto the same postsynaptic neurons is facilitated. We demonstrate that a hierarchical model of cortical processing in the ventral visual system can be trained with continuous transform learning, and highlight differences in the learning of invariant representations to those achieved by trace learning.

Computer Simulation↗
PubMedDate not supplied

Discourse on the transformational leader metanarrative or finding the right person for the job.

Changes in healthcare today have called for leadership styles that place a heavy demand on organizations. Shifts in markets and reimbursement strategies call for new leaders with vision that can guide healthcare organizations to change. In a postmodern era, organizations see the transformational leader as a valuable asset in changing times. This article addresses the foundations of the transformational leader metanarrative and, through a postmodern discourse, deconstructs the concept. The article presents a review of postmodern thought in relation to the metanarrative of leadership and applications to nursing administration.

Humans↗
PubMed2000

Stereoisomeric specificity of the retinoid cycle in the vertebrate retina.

Understanding of the stereospecificity of enzymatic reactions that regenerate the universal chromophore required to sustain vision in vertebrates, 11-cis-retinal, is needed for an accurate molecular model of retinoid transformations. In rod outer segments (ROS), the redox reaction involves all-trans-retinal and pro-S-NADPH that results in the production of pro-R-all-trans-retinol. A recently identified all-trans-retinol dehydrogenase (photoreceptor retinol dehydrogenase) displays identical stereospecificity to that of the ROS enzyme(s). This result is unusual, because photoreceptor retinol dehydrogenase is a member of a short chain alcohol dehydrogenase family, which is often pro-S-specific toward their hydrophobic alcohol substrates. The second redox reaction occurring in retinal pigment epithelium, oxidation of 11-cis-retinol, which is largely catalyzed by abundantly expressed 11-cis-retinol dehydrogenase, is pro-S-specific to both 11-cis-retinol and NADH. However, there is notable presence of pro-R-specific activities. Therefore, multiple retinol dehydrogenases are involved in regeneration of 11-cis-retinal. Finally, the cellular retinaldehyde-binding protein-induced isomerization of all-trans-retinol to 11-cis-retinol proceeds with inversion of configuration at the C(15) carbon of retinol. Together, these results provide important additions to our understanding of retinoid transformations in the eye and a prelude for in vivo studies that ultimately may result in efficient pharmacological intervention to restore and prevent deterioration of vision in several inherited eye diseases.

Alcohol Oxidoreductases↗
PubMed1995

Induced motion may account for the illusory transformation of optic flow fields found by Duffy and Wurtz.

Duffy and Wurtz [(1993) Vision Research, 33, 1481-1490] found an illusory shift in the position of the focus of expansion (FOE) of random dot patterns when planar motion was superimposed on expanding radial motion. Subjects indicated that this illusory shift was in the direction of the planar motion. This is the opposite direction to a true shift in the FOE that is perceived when the planar motion is vector summed with the expanding motion. We account for this illusion by suggesting that planar motion induces opposite motion on the expanding dots which after vector summation produces the illusory shift in the FOE. We use a matching technique with a method of adjustment to measure induced motion and perceived FOE in moving random dot patterns and present the results in support of our assertion.

Humans↗
PubMedDate not supplied

Structure Elucidator: a versatile expert system for molecular structure elucidation from 1D and 2D NMR data and molecular fragments.

StrucEluc is an expert system that allows the computer-assisted elucidation of chemical structures based on the inputs of a series of spectral data including 1D and 2D NMR and mass spectra. The system has been enabled to allow a chemist to utilize fragments stored in a fragment database as well as user-defined fragments submitted by the chemist in the structure elucidation process. The association of fragments in this way has been shown to dramatically speed up the process of structure generation from 2D NMR data and has helped to minimize or eliminate the need for user intervention thereby further enabling the vision of automated elucidation. The use of fragments has frequently transformed very difficult 2D NMR elucidation challenges into easily solvable tasks. A strategy to utilize molecular fragments has been developed and optimized based on specific challenging examples. This strategy will be described here using real world examples. Experience gained by solving more than 150 structure elucidation problems from a variety of literature sources is also reviewed in this work.

Journal Article↗
PubMed2003

Substitution of a non-retinal phospholipase C in Drosophila phototransduction.

The Drosophila norpA gene encodes at least two subtypes of phospholipase C (PLC), one of which is essential for phototransduction and the other is utilized in signalling pathways other than phototransduction. The two subtypes of norpA-PLC differ by 14 amino acids that have been proposed as important for the function of PLC in different signalling pathways. The present study aimed to determine whether norpA subtype II enzyme can functionally substitute for the subtype I enzyme in the phototransduction pathway. We found that the non-retinal norpA-PLC enzyme can substitute for its retinal counterpart, but that there is a reduced rate of repolarization of photoreceptors following intense light stimuli. This reduced repolarization might be due to the inability of a regulatory component being able to interact with the non-retinal norpA-PLC enzyme.

Animals↗
PubMed1995

Phospholipase C rescues visual defect in norpA mutant of Drosophila melanogaster.

Mutations in the norpA gene of Drosophila melanogaster severely affect the light-evoked photoreceptor potential with strong mutations rendering the fly blind. The norpA gene has been proposed to encode phosphatidylinositol-specific phospholipase C (PLC), which enzymes play a pivotal role in one of the largest classes of signaling pathways known. A chimeric norpA minigene was constructed by placing the norpA cDNA behind an R1-6 photoreceptor cell-specific rhodopsin promoter. This minigene was transferred into norpAP24 mutant by P-element-mediated germline transformation to determine whether it could rescue the phototransduction defect concomitant with restoring PLC activity. Western blots of head homogenates stained with norpA antiserum show that norpA protein is restored in heads of transformed mutants. Moreover, transformants exhibit a large amount of measurable PLC activity in heads, whereas heads of norpAP24 mutant exhibit very little to none. Immunohistochemical staining of tissue sections using norpA antiserum confirm that expression of norpA protein in transformants localizes in the retina, more specifically in rhabdomeres of R1-6 photoreceptor cells, but not R7 or R8 photoreceptor cells. Furthermore, electrophysiological analyses reveal that transformants exhibit a restoration of light-evoked photoreceptor responses in R1-6 photoreceptor cells, but not in R7 or R8 photoreceptor cells. This is the strongest evidence thus far supporting the hypothesis that the norpA gene encodes phospholipase C that is utilized in phototransduction.

Animals↗
PubMed1998

Developing our leaders in the future.

The role of the chief executive in a transformed organisation is an extremely challenging one. The development of vision, building a commitment to it and communicating it constantly are key skills for a chief executive. However, the need to build and empower the stakeholders within and outside the organisation to support the changes required to deliver the vision requires leaders who can connect with a wide range of people and build alliances and partnerships to secure organisational success. A passion for understanding human intervention and behaviour is needed to encourage, cajole and drive teams and individuals to own and commit to change and a new direction. This requires leaders who have imagination and creativity--who seek connections and thread them together to create order out of incoherence. These skills are not taught in schools or textbooks, but are probably innate. They are what separate leaders from the rest. These skills need to be developed. A movement towards encouraging experimentation, career transfers and more individuality is needed if capable leaders of the future are to appear.

Administrative Personnel↗
PubMed2004

Involvement of the superior temporal cortex and the occipital cortex in spatial hearing: evidence from repetitive transcranial magnetic stimulation.

The processing of auditory spatial information in cortical areas of the human brain outside of the primary auditory cortex remains poorly understood. Here we investigated the role of the superior temporal gyrus (STG) and the occipital cortex (OC) in spatial hearing using repetitive transcranial magnetic stimulation (rTMS). The right STG is known to be of crucial importance for visual spatial awareness, and has been suggested to be involved in auditory spatial perception. We found that rTMS of the right STG induced a systematic error in the perception of interaural time differences (a primary cue for sound localization in the azimuthal plane). This is in accordance with the recent view, based on both neurophysiological data obtained in monkeys and human neuroimaging studies, that information on sound location is processed within a dorsolateral "where" stream including the caudal STG. A similar, but opposite, auditory shift was obtained after rTMS of secondary visual areas of the right OC. Processing of auditory information in the OC has previously been shown to exist only in blind persons. Thus, the latter finding provides the first evidence of an involvement of the visual cortex in spatial hearing in sighted human subjects, and suggests a close interconnection of the neural representation of auditory and visual space. Because rTMS induced systematic shifts in auditory lateralization, but not a general deterioration, we propose that rTMS of STG or OC specifically affected neuronal circuits transforming auditory spatial coordinates in order to maintain alignment with vision.

Acoustic Stimulation↗
PubMed1989

Appearance of color matchings under various chromatic adaptation conditions.

Three chromatic reference stimuli (red, green, and blue) seen under five adaptation conditions (red, green, blue, achromatic, and darkness) were appearance-matched to subsequent color stimuli seen under achromatic conditions. The experimental results have been used to verify the opponent-color color-vision model and also to check the validity of some of the published transformation equations for predicting the tristimulus values of colors matched to a reference stimulus seen under different chromatic-adaptation conditions. It is concluded that the extant transformation laws are not sufficiently general to be applied to all adaptation conditions.

Adaptation, Ocular↗
PubMed2006

Vision to reality: using patients' voices to develop and improve services.

The NHS is being transformed into a patient-led service requiring radical new ways of gathering patients' views. Sunderland Teaching Primary Care Trust (TPCT) demonstrated its commitment to this change by introducing an action research project in parallel to the introduction of a new nurse-led urgent care team (UCT). The project used a qualitative methodology to collect views from initial UCT service users, patients with chronic obstructive pulmonary disease (COPD). Views were analysed using a thematic framework analysis. A multi-disciplinary action research group considered emerging themes and literature finding solutions and taking action to deliver service improvements. The research provides evidence of the quality of the new service and highlighted a service gap for housebound patients with chronic disease. This needs to be used positively to make improvements for patients in line with the NHS Improvement Plan and long-term conditions agenda.

Community Health Nursing↗
PubMed1978

A general zone theory of color and brightness vision. I. Basic formulation.

A general theory of color and brightness vision, developed from basic principles of the Helmholtz and Hering points-of-view on color vision is presented in a general mathematical form suitable for quantitative analysis. Visual sensation is described by a vector expressed in terms of Hering-like elements for color and brightness which underlie in their spatial-temporal variations the perceptions of form and change. The photic stimulus of vision is recognized to act first and only through photoabsorption producing a Helmholtz-like vector of quantum absorptions. The physiological transformation of the Helmholtz photochemical excitations into the Hering sensation responses is represented as a vector of general operators. The result is a mathematical framework encompassing traditional psychophysical and sensory scaling experiments. The theory is utilized to demonstrate that for many traditional (Class A) psychophysical observations, the physiological operator reduces to a linear (matrix) transformation. For static, uniform, focal stimulation, this reduction is seen to be the basis for earlier specific linear models of color vision. We also illustrate that static intensity-level effects (Bezold-Brücke hue shifts, unique hue invariance) can be modeled from the theory by power, but not logarithmic, intensity-level dependence for the sensation elements.

Color↗
PubMed1992

Quantitative evaluation of flicker ERG waveforms in low vision patients.

Flicker electroretinograms (FERGs) of 30Hz were recorded for 46 eyes (23 cases) of low vision, including 38 eyes with retinal diseases and 8 eyes with optic neuropathy, and for 39 normal eyes (33 cases). The recorded wave forms were analysed by two methods: (1) to measure the peak to peak amplitude and pseudophase directly, and (2) to measure the amplitude and phase of fundamental response component (30Hz) by discrete Fourier transform (DFT). The abnormality ratios between these two methods were compared in low vision patients. All the eyes with optic neuropathy showed normal flicker ERG, the abnormality ratios of directly measured amplitude and the amplitude by DFT were 60.9% in 28 eyes and 71.7% in 33 eyes with retinal diseases respectively, the difference being no statistically significant (P > 0.05), while the abnormality ratios of directly measured pseudo-phase and the phase by DFT were 60.9% in 28 eyes and 82.6% in 38 eyes with retinal diseases respectively, the difference being statistically significant (P < 0.05). These results suggest that FERGs are an objective way to test the retinal function, and DFT for flicker ERGs are more useful in quantitative evaluation of the retinal function.

Adolescent↗
PubMed2001

Links between vision and somatosensation. Vision can improve the felt position of the unseen hand.

During reaching movements, sensory signals must be transformed into appropriate motor commands. Anatomical, electrophysiological, and neuropsychological evidence suggest that there is no single, supramodal map of space that is used to guide reaching. Instead, movements appear to be planned and controlled within multiple coordinate systems, each one attached to a different body part. Recent neuropsychological investigations demonstrating that somatosensory impairments can be ameliorated by visual cues, and visual impairments by proprioceptive cues, have been interpreted as evidence that arm-centered representations may exist in humans. A critical difference between the findings obtained in the monkey and in humans, however, is that in the latter case, vision of the limb appears be critical for visual somatosensory binding. Here, we report a case study of a patient (C.T.) recovering from unilateral somatosensory impairment, including tactile extinction, who executed reaches toward visually defined or proprioceptively defined locations. We demonstrate that when the target location of a reach was defined proprioceptively, by passively positioning our patient's impaired hand beneath the table surface, vision of the workspace immediately adjacent to the unseen hand dramatically increased the endpoint accuracy of her reaching movements, even though such cues could not possibly signal the position of the target directly.

Aged↗
PubMed2005

A method and software for segmentation of anatomic object ensembles by deformable m-reps.

Deformable shape models (DSMs) comprise a general approach that shows great promise for automatic image segmentation. Published studies by others and our own research results strongly suggest that segmentation of a normal or near-normal object from 3D medical images will be most successful when the DSM approach uses (1) knowledge of the geometry of not only the target anatomic object but also the ensemble of objects providing context for the target object and (2) knowledge of the image intensities to be expected relative to the geometry of the target and contextual objects. The segmentation will be most efficient when the deformation operates at multiple object-related scales and uses deformations that include not just local translations but the biologically important transformations of bending and twisting, i.e., local rotation, and local magnification. In computer vision an important class of DSM methods uses explicit geometric models in a Bayesian statistical framework to provide a priori information used in posterior optimization to match the DSM against a target image. In this approach a DSM of the object to be segmented is placed in the target image data and undergoes a series of rigid and nonrigid transformations that deform the model to closely match the target object. The deformation process is driven by optimizing an objective function that has terms for the geometric typicality and model-to-image match for each instance of the deformed model. The success of this approach depends strongly on the object representation, i.e., the structural details and parameter set for the DSM, which in turn determines the analytic form of the objective function. This paper describes a form of DSM called m-reps that has or allows these properties, and a method of segmentation consisting of large to small scale posterior optimization of m-reps. Segmentation by deformable m-reps, together with the appropriate data representations, visualizations, and user interface, has been implemented in software that accomplishes 3D segmentations in a few minutes. Software for building and training models has also been developed. The methods underlying this software and its abilities are the subject of this paper.

Algorithms↗