Abstract: Methods and systems for generating saliency models are discussed. Saliency models can be applied to visual scenes to generate predictions on which locations in the visual scenes are fixation locations and which locations are nonfixation locations. Saliency models are learned from fixation data on the visual scenes obtained from one or more subjects.

Abstract: An online real-time (ORT) system and method implementing such system for real-time prediction of one of two actions or classes of action are described. Such actions are detected by corresponding transducers configured to translate the actions to time varying amplitude signals.

Abstract: Detection of image salience in a visual display of an image. The image is analyzed at multiple spatial scales and over multiple feature channels to determine the likely salience of different portions of the image. One application for the system is in an advertising context. The detection may be improved by second order statistics, e.g. mean and the standard deviations of different image portions relative to other portions. Different edges may be considered as being extended edges by looking at the edges over multiple spatial scales. One set of feature channels can be optimized for use in moving images, and can detect motion or flicker. The images can be obtained over multiple spectral ranges the user can be instructed about how to maximize the saliency. This can be applied to automatically evaluate and optimize sales or advertisement displays.

Abstract: Detection of image salience in a visual display of an image. The image is analyzed at multiple spatial scales and over multiple feature channels to determine the likely salience of different portions of the image. One application for the system is in an advertising context. The detection may be improved by second order statistics, e.g. mean and the standard deviations of different image portions relative to other portions. Different edges may be considered as being extended edges by looking at the edges over multiple spatial scales. One set of feature channels can be optimized for use in moving images, and can detect motion or flicker. The images can be obtained over multiple spectral ranges the user can be instructed about how to maximize the saliency. This can be applied to automatically evaluate and optimize sales or advertisement displays.

Abstract: Detection of image salience in a visual display of an image. The image is analyzed at multiple spatial scales and over multiple feature channels to determine the likely salience of different portions of the image. One application for the system is in an advertising context. The detection may be improved by second order statistics, e.g. mean and the standard deviations of different image portions relative to other portions. Different edges may be considered as being extended edges by looking at the edges over multiple spatial scales. One set of feature channels can be optimized for use in moving images, and can detect motion or flicker. The images can be obtained over multiple spectral ranges the user can be instructed about how to maximize the saliency. This can be applied to automatically evaluate and optimize sales or advertisement displays.

Abstract: Methods and systems for generating saliency models are discussed. Saliency models can be applied to visual scenes to generate predictions on which locations in the visual scenes are fixation locations and which locations are nonfixation locations. Saliency models are learned from fixation data on the visual scenes obtained from one or more subjects.

Abstract: Detection of image salience in a visual display of an image. The image is analyzed at multiple spatial scales and over multiple feature channels to determine the likely salience of different portions of the image. One application for the system is in an advertising context. The detection may be improved by second order statistics, e.g. mean and the standard deviations of different image portions relative to other portions. Different edges may be considered as being extended edges by looking at the edges over multiple spatial scales. One set of feature channels can be optimized for use in moving images, and can detect motion or flicker. The images can be obtained over multiple spectral ranges the user can be instructed about how to maximize the saliency. This can be applied to automatically evaluate and optimize sales or advertisement displays.

Abstract: The present invention relates to a system and method for attentional selection. More specifically, the present invention relates to a system and method for the automated selection and isolation of salient regions likely to contain objects, based on bottom-up visual attention, in order to allow unsupervised one-shot learning of multiple objects in cluttered images.

Abstract: Detection of image salience in a visual display of an image. The image is analyzed at multiple spatial scales and over multiple feature channels to determine the likely salience of different portions of the image. One application for the system is in an advertising context. The detection may be improved by second order statistics, e.g. mean and the standard deviations of different image portions relative to other portions. Different edges may be considered as being extended edges by looking at the edges over multiple spatial scales. One set of feature channels can be optimized for use in moving images, and can detect motion or flicker. The images can be obtained over multiple spectral ranges the user can be instructed about how to maximize the saliency. This can be applied to automatically evaluate and optimize sales or advertisement displays.

Abstract: An integrated circuit that computes the velocity of a visual stimulus moving between two photoreceptor locations is disclosed. In its most basic version, the circuit comprises two temporal edge detectors with photoreceptors, two pulse-shaping circuits, and one motion circuit on a single silicon chip. Velocity is computed from the signed time delay of the appearance of an image feature at the two photoreceptor locations. Specifically, each temporal edge detector de tects a rapid irradiance transient at its photoreceptor location and converts it into a short current spike. This current spike is transformed into two different voltage pulses, a fast pulse and a slowly-decaying pulse, by the pulse-shaping circuit that is coupled to the temporal edge detector.

Abstract: An integrated circuit that computes the velocity of a visual stimulus moving between two photoreceptor locations is disclosed. In its most basic version, the circuit comprises two temporal edge detectors with photoreceptors, two pulse-shaping circuits, and one motion circuit on a single silicon chip. Velocity is computed from the signed time delay of the appearance of an image feature at the two photoreceptor locations. Specifically, each temporal edge detector detects a rapid irradiance transient at its photoreceptor location and converts it into a short current spike. This current spike is transformed into two different voltage pulses, a fast pulse and a slowly-decaying pulse, by the pulse-shaping circuit that is coupled to the temporal edge detector.

Abstract: An integrated circuit that computes the velocity of a visual stimulus moving between two photoreceptor locations is disclosed. In its most basic version, the circuit comprises two temporal edge detectors with photoreceptors, two pulse-shaping circuits, and one motion circuit on a single silicon chip. Velocity is computed from the signed time delay of the appearance of an image feature at the two photoreceptor locations. Specifically, each temporal edge detector detects a rapid irradiance transient at its photoreceptor location and converts it into a short current spike. This current spike is transformed into two different voltage pulses, a fast pulse and a slowly-decaying pulse, by the pulse-shaping circuit that is coupled to the temporal edge detector.

Abstract: An integrated circuit that computes the velocity of a visual stimulus moving between two photoreceptor locations is disclosed. In its most basic version, the circuit comprises two temporal edge detectors with photoreceptors, two pulse-shaping circuits, and one motion circuit on a single silicon chip. Velocity is computed from the signed time delay of the appearance of an image feature at the two photoreceptor locations. Specifically, each temporal edge detector detects a rapid irradiance transient at its photoreceptor location and converts it into a short current spike. This current spike is transformed into two different voltage pulses, a fast pulse and a slowly-decaying pulse, by the pulse-shaping circuit that is coupled to the temporal edge detector.

Abstract: An integrated circuit that computes the velocity of a visual stimulus moving between two photoreceptor locations is disclosed. In its most basic version, the circuit comprises two temporal edge detectors with photoreceptors, two pulse-shaping circuits, and one motion circuit on a single silicon chip. Velocity is computed from the signed time delay of the appearance of an image feature at the two photoreceptor locations. Specifically, each temporal edge detector detects a rapid irradiance transient at its photoreceptor location and converts it into a short current spike. This current spike is transformed into two different voltage pulses, a fast pulse and a slowly-decaying pulse, by the pulse-shaping circuit that is coupled to the temporal edge detector.

Abstract: An analog VLSI circuit having a network of resistive elements and switches is provided for discriminating between an object and the background in a sensor image of a scene. A sensor system captures two-dimensional visual images and identifies edges of objects. The output of the imaging system is mapped onto the switched resistive grid. The presence of object edges causes switches to open at corresponding grid points in the resistive network. The periphery of the grid is connected to a first voltage, while a central point of the grid is connected to a second voltage. An edge map contour of open switches surrounding the central point causes the voltage at every point within the contour to rise to the second voltage, while every grid point outside the contour settles to the first voltage. Thus, the object enclosed by the edge map contour is labeled or tagged with the second voltage and the surrounding background is characterized by the first voltage.

Abstract: A computer vision edge-detection circuit uses two independent resistive networks to smooth the voltages supplied by an array of logarithmic photoreceptors. The voltages on the two networks are substracted and exclusive-or circuitry is used to detect zero-crossings. In order to facilitate thresholding of the edges, an additional current is computed at each node indicating the strength of the zero-crossing. This is particularly important for robust real-world performance where there will be many small zero-crossings induced by noise. Implementation of the present invention using analog VLSI techniques makes it feasible, and, in fact, practical to provide a large plurality of such photoreceptors on a single integrated circuit chip. Such chips can provide image acquisition and edge-detection with noise elimination in one chip in real time.

Abstract: This invention is primarily a "resistive fuse" circuit, being a hardware circuit that explicitly implements either analog or binary line processes in a controlled fashion.