Biomedical subjects

M S Landy

Publications and source records attributed to M S Landy.

11 recordsLinked to original sources

PubMed1991

Texture segregation and orientation gradient.

Rapid texture segregation is examined using filtered noise textures. The stimuli consist of a foreground region of filtered noise with one dominant texture orientation against a background region with a different dominant orientation. Shape discrimination of the foreground region is measured as a function of the difference in orientation between the two regions (delta theta), the distance over which the dominant orientation rotates from the background to the foreground value (delta chi), and the dominant spatial frequency of the textures (f). Performance declines with smaller delta theta, larger delta chi, and lower f. These effects are partially independent of viewing distance, which implies that it is the relative or object spatial frequency, not retinal spatial frequency, which determines performance in this task. We present a model consisting of channels tuned for orientation and spatial frequency which compute local oriented energy, followed by (texture) edge detection and a cross-correlator which performs the shape discrimination. Monte Carlo simulations of this model are in accord with the degradation in performance with increased delta chi and decreased delta theta.

Humans↗

PubMed1991

The kinetic depth effect and optic flow--II. First- and second-order motion.

We use a difficult shape identification task to analyze how humans extract 3D surface structure from dynamic 2D stimuli--the kinetic depth effect (KDE). Stimuli composed of luminous tokens moving on a less luminous background yield accurate 3D shape identification regardless of the particular token used (either dots, lines, or disks). These displays stimulate both the 1st-order (Fourier-energy) motion detectors and 2nd-order (nonFourier) motion detectors. To determine which system supports KDE, we employ stimulus manipulations that weaken or distort 1st-order motion energy (e.g. frame-to-frame alternation of the contrast polarity of tokens) and manipulations that create microbalanced stimuli which have no useful 1st-order motion energy. All manipulations that impair 1st-order motion energy correspondingly impair 3D shape identification. In certain cases, 2nd-order motion could support limited KDE, but it was not robust and was of low spatial resolution. We conclude that 1st-order motion detectors are the primary input to the kinetic depth system. To determine minimal conditions for KDE, we use a two frame display. Under optimal conditions, KDE supports shape identification performance at 63-94% of full-rotation displays (where baseline is 5%). Increasing the amount of 3D rotation portrayed or introducing a blank inter-stimulus interval impairs performance. Together, our results confirm that the human KDE computation of surface shape uses a global optic flow computed primarily by 1st-order motion detectors with minor 2nd-order inputs. Accurate 3D shape identification requires only two views and therefore does not require knowledge of acceleration.

Depth Perception↗

PubMed1991

Nonadditivity of masking by narrow-band noises.

Characterization of the visual system as a linear system has many consequences. One property implied by such a characterization is that signal-to-noise ratio at threshold is constant, so that the contrast energy of a sinusoidal grating at threshold depends on the effective noise passed by the filter used to detect the target. When the contrast energy in an external noise is sufficiently high, the contribution of internal noise may be conveniently ignored. In such circumstances, the effectiveness of a given noise is measured by its ability to mask the target. One consequence of the linearity assumption is that if the energy of the effective noise is increased by a factor k, then threshold energy of the signal is increased by the same amount. The contrast energy at threshold in the presence of a masker created by adding two maskers should be the sum of the threshold energies in the individual maskers. We have tested this hypothesis for spectrally nonoverlapping maskers. We find that the contrast energy required to detect the target in the presence of the combined maskers is much greater than the sum of the threshold energies for the two maskers. This "excess masking" violates the linearity assumption. On the other hand, when maskers do have spectral overlap, no excess masking is found.

Contrast Sensitivity↗

PubMed1990

Applications of the EVE software for visual modeling.

EVE, the Early Vision Emulation software, is a system for the stimulation of early visual processing. EVE has the ability to carry out the operations of a wide variety of models of spatial vision, motion detection and processing, and spatial sampling. We introduce the EVE software and illustrate some of its applications for models of pattern detection, pattern discrimination, and motion detection.

Humans↗

PubMed1990

Intelligent temporal subsampling of American Sign Language using event boundaries.

How well can a sequence of frames be represented by a subset of the frames? Video sequences of American Sign Language (ASL) were investigated in two modes: dynamic (ordinary video) and static (frames printed side by side on the display). An activity index was used to choose critical frames at event boundaries, times when the difference between successive frames is at a local minimum. Sign intelligibility was measured for 32 experienced ASL signers who viewed individual signs. For full gray-scale dynamic signs activity-index subsampling yielded sequences that were significantly more intelligible than when every mth frame was chosen. This result was even more pronounced for static images. For binary images, the relative advantage of activity subsampling was smaller. We conclude that event boundaries can be defined computationally and that subsampling from event boundaries is better than choosing at regular intervals.

Adolescent↗

PubMed1990

How to study the kinetic depth effect experimentally.

Sperling, Landy, Dosher, and Perkins (1989) proposed an objective 3D shape identification task with 2D artifactual cues removed and with full feedback (FB) to the subjects to measure KDE and to circumvent algorithmically equivalent KDE-alternative computations and artifactual non-KDE processing. (1) The 2D velocity flow-field was necessary and sufficient for true KDE. (2) Only the first-order (Fourier-based) perceptual motion system could solve our task because the second-order (rectifying) system could not simultaneously process more than two locations. (3) To ensure first-order motion processing, KDE tasks must require simultaneous processing at more than two locations. (4) Practice with FB is essential to measure ultimate capacity (aptitude) and, thereby, to enable comparisons with ideal observers. Experiments without FB measure ecological achievement--the ability of subjects to extrapolate their past experience to the current stimuli.

Algorithms↗

PubMed1989

Kinetic depth effect and optic flow--I. 3D shape from Fourier motion.

Fifty-three different 3D shapes were defined by sequences of 2D views (frames) of dots on a rotating 3D surface. (1) Subjects' accuracy of shape identifications dropped from over 90% to less than 10% when either the polarity of the stimulus dots was alternated from light-on-gray to dark-on-gray on successive frames or when neutral gray interframe intervals were interposed. Both manipulations interfere with motion extraction by spatio-temporal (Fourier) and gradient first-order detectors. Second-order (non-Fourier) detectors that use full-wave rectification are unaffected by alternating-polarity but disrupted by interposed gray frames. (2) To equate the accuracy of two-alternative forced-choice (2AFC) planar direction-of-motion discrimination in standard and polarity-alternated stimuli, standard contrast was reduced. 3D shape discrimination survived contrast reduction in standard stimuli whereas it failed completely with polarity-alternation even at full contrast. (3) When individual dots were permitted to remain in the image sequence for only two frames, performance showed little loss compared to standard displays where individual dots had an expected lifetime of 20 frames, showing that 3D shape identification does not require continuity of stimulus tokens. (4) Performance in all discrimination tasks is predicted (up to a monotone transformation) by considering the quality of first-order information (as given by a simple computation on Fourier power) and the number of locations at which motion information is required. Perceptual first-order analysis of optic flow is the primary substrate for structure-from-motion computations in random dot displays because only it offers sufficient quality of perceptual motion at a sufficient number of locations.

Depth Perception↗

PubMed1989

Ratings of kinetic depth in multidot displays.

Subjects saw kinetic depth displays whose shape (sphere or cylinder) was defined by luminous dots distributed randomly on the surface or in the volume of the object. Subjects rated perceived 3-D depth, rigidity, and coherence. Despite individual differences, all 3 ratings increased with the number of dots. Dots in the volume yielded ratings equal to or greater than surface dots. Each rating varied with 3 of 4 factors (shape, distribution, numerosity, and perspective), but the ratings either between trials or between conditions were often uncorrelated. Object shape affected rigidity but not depth ratings. Veridically perceived polar displays had slightly lower rigidity but higher depth ratings than parallel projection displays. (Reversed polar displays were always grossly nonrigid.) The interaction of ratings and stimulus parameters requires theories and experiments in which different KDE ratings are not treated interchangeably.

Attention↗

PubMed1989

Kinetic depth effect and identification of shape.

We introduce an objective shape-identification task for measuring the kinetic depth effect (KDE). A rigidly rotating surface consisting of hills and valleys on an otherwise flat ground was defined by 300 randomly positioned dots. On each trial, 1 of 53 shapes was presented; the observer's task was to identify the shape and its overall direction of rotation. Identification accuracy was an objective measure, with a low guessing base rate of the observer's perceptual ability to extract 3D structure from 2D motion via KDE. (1) Objective accuracy data were consistent with previously obtained subjective rating judgments of depth and coherence. (2) Along with motion cues, rotating real 3D dot-defined shapes inevitably produced a cue of changing dot density. By shortening dot lifetimes to control dot density, we showed that changing density was neither necessary nor sufficient to account for accuracy; motion alone sufficed. (3) Our shape task was solvable with motion cues from the 6 most relevant locations. We extracted the dots from these locations and used them in a simplified 2D direction-labeling motion task with 6 perceptually flat flow fields. Subjects' performance in the 2D and 3D tasks was equivalent, indicating that the information processing capacity of KDE is not unique. (4) Our proposed structure-from-motion algorithm for the shape task first finds relative minima and maxima of local velocity and then assigns 3D depths proportional to velocity.

Adult↗

PubMed1989

Depth interpolation with sparse disparity cues.

The interpolation of stereoscopic depth given only sparse disparity information was investigated. The basic stimulus was a rectangle with zero disparity at one edge, and 20 or 30 min visual angle disparity at the other. The depth assigned to the ambiguous intervening locations was measured by means of a small briefly-flashed binocular comparison spot. For a stimulus consisting of a uniform rectangle presented on a background of random dots with zero disparity, interpolated depth was greater for a high mean contrast between rectangle and background than for a low mean contrast. Relative to a linear interpolation between the edges, a larger difference in edge disparity resulted in poorer depth interpolation. Depth interpolation based on rivalrous information was examined by filling the stimulus rectangle with narrow-band filtered noise which was uncorrelated between the two eyes. Four different passbands which were matched in apparent contrast were investigated. The results demonstrate that the rivalrous low-spatial-frequency content was resistant to interpolation; rivalrous high spatial frequencies did not interfere with depth interpolation. High-spatial-frequency stimuli yielded a percept similar to the uniform-field condition, whereas low-spatial-frequency stimuli lay in a depth plane near or even behind the background. In the latter case a transparent plane was perceived which was linearly interpolated between the two edges, and which floated above the rivalrous noise.

Attention↗

PubMed1987

Parallel model of the kinetic depth effect using local computations.

This paper defines a new model for the kinetic depth effect for multidot stimuli. The calculation is performed in a cooperative-competitive network, described as a relaxation labeling process. The process involves a local iterative computation to meet best the constraints indicated by image cues to depth. Given a constraint that prefers interdot distances in three dimensions to remain constant (local rigidity), the model becomes a local parallel computation of the Ullman incremental-rigidity scheme. Several simulations of the model are described, including some in which additional cues are combined with the changing-dot-position cue.

Computer Simulation↗