391 results about "Left eye" patented technology

Disclosed in the present invention is a salient contour matching-based method for target measurement in a severe imaging environment. The method specifically comprises: (1) acquiring a binocular image of a target; (2) establishing a global-local joint constraint-based background light estimation model, and removing a scattering effect of a medium in an imaging environment to obtain a restored left eye image and a restored right eye image; (3) learning an original image, and on the basis of a residual between a network reconstructed image and the original image, obtaining target localization prediction maps of the left eye image and the right eye image; and (4) respectively extracting contour lines of the target in the left eye image and the right eye image, constructing feature matching descriptors of contour points, performing stereo matching on the two sets of contour lines by minimizing matching cost, and performing three-dimensional reconstruction on the contour lines in light of calibrated intrinsic and extrinsic parameters to complete the measurement of a key size. According to the present invention, the key sizes of different targets in a severe environment can be accurately measured, thereby providing an effective solution for the problem of measuring the sizes of targets in a severe environment.

The subject invention pertains to an artificial intelligence-aided classification system for Alzheimer's disease (AD) screening from retinal photographs, which includes a three-in-one deep learning (DL)-based pre-diagnosis module for image assessment and a DL-based AD classification module with additional heatmaps for visualization. For the AD classification module, three kinds of DL models predict the AD-dementia / non-demented probabilities from three directions with both eyes' four images (Direction-1), both eyes' four images combing demographical information (Direction-2), and a single eye's two images (Direction-3), respectively. For the pre-diagnosis image assessment, embodiments provide one pre-processing model, with three additional models for each of the three classification tasks, including image-quality (e.g., gradable or ungradable), field-of-view (e.g., macula-centered or optic nerve head-centered), and laterality-of-the-eye (e.g., right or left eye). Also included is a cloud-based web application that can output the pre-diagnosis image assessment and a simple binary AD-dementia / non-demented classification based on retinal photographs.

Various examples disclosed herein maintain stereo consistency in extended reality (XR) environments when receiving content data with a content frame depicting both a left eye portion of a content item corresponding to a first left eye viewpoint of the content item within the XR environment, and a right eye portion of the content item corresponding to a first right eye viewpoint of the content item within the XR environment. Stereo consistency may be maintained by determining to use an adjusted version of the content frame to provide a view of the content item wherein a left eye view from a second left eye viewpoint is different than the first left eye viewpoint and a right eye view from a second right eye viewpoint is different than the first right eye viewpoint, and presenting the left eye view and the right eye view based on the content frame and the adjustment.

This invention belongs to the field of facial expression recognition and provides a micro-expression recognition method based on convolutional neural networks and optical flow features, which is used to improve the adaptability and accuracy of micro-expression recognition. The method first constructs a face detection module to perform face detection on each image frame in a video segment, outputs facial key point information, and extracts facial regions of interest (ROIs) based on the facial key point information: nose, mouth, left eye and eyebrow, right eye and eyebrow. Then, optical flow information and optical strain information are introduced to characterize the spatiotemporal information of facial movement and the intensity information of facial deformation. The optical flow eigenvalue of each image frame is calculated, and a binary search algorithm is used to search for the image frame with the maximum optical flow eigenvalue, which is used as the vertex frame for micro-expression recognition. Finally, a micro-expression recognition model based on a convolutional neural network is constructed, which uses the horizontal and vertical components of the optical flow and optical strain of the vertex frame as input. The micro-expression recognition model outputs a micro-expression category prediction result.

The invention provides a human-computer interaction method based on eye movement tracking, and the method is characterized in that the method comprises the steps: S1, obtaining an original image, and carrying out the preprocessing of the original image; s2, generating a depth information image of the preprocessed original image, and performing facial feature detection and interception on the preprocessed original image to obtain a facial feature image; s3, performing eye feature detection and interception on the facial feature image to respectively obtain a left eye feature image and a right eye feature image, and obtaining user feature information according to the facial feature image; s4, processing the facial feature image, the left eye feature image and the right eye feature image by using a deep learning model, and calculating a fixation point in combination with the depth information image; and S5, outputting user features and gaze information by combining the gaze point and the user feature information, and executing a corresponding program in response to the interactive action. According to the invention, accurate calculation of the depth information of each pixel point in the image is realized, and the depth information graph completely corresponding to the input image is generated.

The invention belongs to the technical field of face recognition, and provides a face recognition system and an anesthesia cabinet using the same, and the system comprises an image collection module which is used for obtaining the face image data of a to-be-stored user through face detection; the feature extraction module is used for acquiring a planar graph of the facial image of the to-be-stored user through the facial image data of the to-be-stored user; the calculation module is used for subdividing the left eye region and the right eye region of the user to be stored through a loop subdivision algorithm; the face recognition module is used for performing face detection through the recognition module to obtain face image data of the to-be-detected user when the to-be-detected user needs to open the anesthetic cabinet; and the judgment module is used for judging whether the anesthetic cabinet is opened. According to the face recognition system provided by the embodiment of the invention, face information can be directly and accurately recognized through eyes in a medical environment.

An eye tracking system can include a first camera configured to capture a first plurality of visual data of a right eye at a first sampling rate. The system can include a second camera configured to capture a second plurality of visual data of a left eye at a second sampling rate. The second plurality of visual data can be captured during different sampling times than the first plurality of visual data. The system can estimate, based on at least some visual data of the first and second plurality of visual data, visual data of at least one of the right or left eye at a sampling time during which visual data of an eye for which the visual data is being estimated are not being captured. Eye movements of the eye based on at least some of the estimated visual data and at least some visual data of the first or second plurality of visual data can be determined.

To provide a wearable device, a system, and a display method that control the display on the basis of the results of detecting the rotation of the right eye and the rotation of the left eye.SOLUTION: The wearable device is configured to determine whether a user is looking at a first object at a first distance in the real world or looking at a second object at a second distance which is different from the first distance on the basis of the angle of convergence detected by a detector, display a first augmented reality image on a display when the user is determined to be looking at the first object, transmit a request signal to an electronic device when the user is determined to be looking at the second object, and display a second virtual reality image transmitted from the electronic device in response to the request signal on the display.SELECTED DRAWING: Figure 4

The invention discloses a mechanical arm vision distance measurement method and a distance measurement system thereof, and relates to the technical field of intelligent vision perception, a camera is installed on a mechanical arm to collect images, and through an optimization matching algorithm, the images shot by the left eye and the right eye are rapidly compared to generate a clear and accurate disparity map; the motion state of the mechanical arm is detected in real time through a vibration compensation mechanism, a shot image is corrected, and it is ensured that the distance measuring process is stable and reliable; and in combination with the corrected disparity map, the accurate distance from the target object to the camera is obtained by utilizing a depth calculation formula, and the result is transmitted to a control system of the mechanical arm. The depth value is fed back to the mechanical arm controller in real time through depth calculation and linkage with a control system, the precision of dynamic path planning and action execution of the mechanical arm is improved, and the mechanical arm shows higher anti-interference capacity in a dynamic vibration environment.

The integrated VR glasses provided by the invention comprise a glasses frame, the glasses frame comprises a glasses frame and glasses legs, and the glasses legs are rotatably connected to the glasses frame; the display module is embedded in the mirror frame and used for displaying pictures watched by the left eye and the right eye, the display module comprises a first display screen and a second display screen, the first display screen is arranged at the left mirror position of the mirror frame, the second display screen is arranged at the right mirror position of the mirror frame, the first display screen displays the pictures watched by the left eye, and the second display screen displays the pictures watched by the right eye; the nose pad structure is detachably mounted on the glasses frame; the earphones are fixed on the glasses legs, are in communication connection with the display module and are used for outputting audio information; the posture sensing module is arranged on the glasses frame and used for obtaining user action information, converting the action information into control information and sending the control information to the display module. According to the invention, the in-ear earphones are fixed with the glasses legs through the universal arms, and the universal arms can enable the earphones to stay at any position, so that the in-ear earphones can be ensured to adapt to all users.

Provided in the present application are an intraocular pressure measurement apparatus, a method for identifying left and right eyes, and a storage medium. The intraocular pressure measurement apparatus comprises: a main body base, a transmitter, at least one receiver, and a controller, wherein the main body base comprises a probe; the transmitter is arranged on the main body base, and is used for transmitting a test signal, and the test signal is adjusted to be propagated towards a preset angle range; the receiver is arranged on the main body base, and is used for receiving a reflected signal of the test signal; the controller is connected to both the transmitter and the receiver, and the transmitter and the receiver are located on the same side of the probe; and the controller is configured to determine, according to the signal strength of the reflected signal, whether the eye currently under measurement is a left eye or a right eye, or when the test signal carries a code, the controller is configured to determine, according to the code carried in the reflected signal, whether the eye currently under measurement is a left eye or a right eye.

A method of operating an electronic device to mitigate double vision is provided. The method can include using front-facing cameras to acquire passthrough content, using a graphics renderer to generate virtual content, mitigating double vision or improving vergence comfort by modifying at least the passthrough content, merging the passthrough content with the virtual content, and displaying the merged content. The passthrough content can be modified by selectively shifting or transforming left eye passthrough content or right eye passthrough content. The virtual content can be rendered based on a similar type of modification applied to the passthrough content. The virtual content can also be modified by selectively shifting or transforming left eye virtual content or right eye virtual content.

The invention discloses a light-weight heat map stereo matching method and device, which are used for solving the technical problem that the precision of a final parallax result is poor due to the fact that information is easy to lose when a tiny depth difference is captured by the existing light-weight heat map stereo matching method. The method comprises the following steps: acquiring a left eye thermal image and a right eye thermal image, and inputting a preset thermal image stereo matching model (including multi-scale fusion feature extraction, edge feature extraction, an enhanced wavelet attention module, a plurality of cost aggregation residual blocks and a convolution module); a multi-scale module extracts features, generates a left eye and right eye fusion feature map, and constructs an initial cost body; the edge module outputs a left-eye edge probability feature graph, and an intermediate cost body is obtained by combining the initial cost body; the wavelet attention module outputs an attention weight according to the left eye fusion feature map; the cost aggregation residual block is multiplied by the middle cost body according to the weight, and an initial disparity map is generated through disparity regression; and the convolution module combines the left-eye edge probability feature map for refining, and finally generates a target disparity map.

According to an aspect, a system includes a first image source for a left eye and a second image source for a right eye. The system further includes at least one window in a first optical path for the first image source or a second optical path for the second image source, the at least one window configured to align first pixels from the first image source with second pixels from the second image source.

The present invention provides a driver fatigue state detection method and system, comprising: collecting a driver's facial image; inputting the driver's facial image into a deep learning fast face detection algorithm to obtain left eye center coordinates, right eye center coordinates and binocular distance; calculating a left eye circumscribed rectangular frame and a right eye circumscribed rectangular frame according to the left eye center coordinates, right eye center coordinates and binocular distance; determining an eye rectangular frame according to the left eye circumscribed rectangular frame and the right eye circumscribed rectangular frame; inputting the eye rectangular frame into a pupil detection model to obtain pupil area coordinates; calculating an adaptive threshold according to the pupil area coordinates; calculating a mean aspect ratio within a preset time period; comparing the mean aspect ratio with the adaptive threshold, and determining the current state of the driver according to the comparison result; and solving the problems of inaccurate fatigue detection and weak robustness caused by factors such as individual differences of drivers or complex lighting in the prior art.

A method of assessing a health condition of a mammalian subject includes using a pupilometer to take three or more sequential measurements over a period of time of the left eye and the right eye of the subject and determining the NPI of each eye at each measurement, wherein the subject is hospitalized as a result of stroke, TBI, acute brain injury or other neurological impairment. The method also includes displaying on a display of the pupilometer a comparison between the NPI of the left and the NPI of right eye at each of the three or more sequential measurements, determining whether the subject exhibits NPI differential persistence, and assessing the subject with having an increased risk of a bad neurological or health outcome if the patient exhibits NPI differential persistence.

A head-mounted display according to an embodiment includes a combiner configured to combine display light for forming a display image and outside light from in front of a user wearing the head-mounted display, and a divider part arranged between a space in front of a left eye of the user and a space in front of a right eye, and configured to diffusely reflect the display light.

To provide a display device that can prevent wrong operation to easily switch settings.SOLUTION: A display device 100 has: a first eyepiece optical system that projects images to one of the right eye and the left eye of a user H; a second eyepiece optical system that projects images to the other of the right eye and the left eye of the user; a detection unit that detects a posture of the display device; and a control unit that controls the images to be projected. The control unit controls the images to be projected on the basis of a result of detection performed by the detection unit, so as to project first images in a first posture in which the user looks into the first eyepiece optical system with the right eye and looks into the second eyepiece optical system with the left eye, and project second images different from the first images in a second posture in which the user looks into the second eyepiece optical system with the right eye and looks into the first eyepiece optical system with the left eye.SELECTED DRAWING: Figure 1

A virtual image display device includes a first display device for a right eye, a second display device for a left eye, a lens barrel or the like serving as an assembly member configured to assemble the first display device and the second display device, and first and second correction units serving as post-assembly adjustment devices configured to adjust a display state of the first display device and the second display device assembled by the assembly member.

A method includes receiving scene image data, comprising left and right images for each eye. Through an eye tracking module, the method includes determining the user's gaze direction or eye vergence. Using this information, the method includes identifying an object in the scene that the user is focusing on, and determining, using a depth estimation module, a left depth from the left eye to the object and a right depth from the right eye to the object. Further, based on the computed left and right depths, the method includes generating, for the left and right eye, constant left and right depth meshes, generating a left and right output images by projecting the left and right images on the corresponding constant left and right depth meshes. Additionally, the method includes displaying the left output image for the left eye and displaying the right output image for the right eye.

The invention relates to an in-situ three-dimensional reconstruction method for a weak-texture narrow inner cavity of an aero-engine, and relates to the technical field of reverse modeling, and the method comprises the steps: obtaining a left-eye visual angle sequence image and a right-eye visual angle sequence image, collected based on a binocular borescope, of a key part in the inner cavity of the aero-engine, constructing local point cloud data of the key component according to the left eye visual angle sequence image and the right eye visual angle sequence image; determining feature matching vectors among the key components according to the left eye visual angle sequence image, and determining feature point matching pairs among the key components according to the feature matching vectors; determining an essential matrix of a left side borescope in the binocular borescope according to the feature point matching pairs, and determining pose state parameters of the left side borescope according to the essential matrix of the left side borescope; and performing in-situ three-dimensional reconstruction on the weak-texture narrow inner cavity of the aero-engine according to the pose state parameters and the local point cloud data. The accuracy of the three-dimensional reconstruction result is improved.

The invention provides a brain surgery image analysis method based on a computer vision technology, and relates to the technical field of computer vision, and the method comprises the steps: collecting a three-dimensional binocular image in a brain surgery; determining visual difference between a left eye and a right eye in the binocular camera, generating a depth map space of a three-dimensional binocular image, and further extracting a visual missing area in the brain surgery; determining an obstacle shielding domain of an obstacle in the binocular camera according to the depth map space, and performing light stripe tracking through the obstacle shielding domain and a relative position between a left eye and a right eye in the binocular camera to obtain light stripe missing information in the brain surgery; and determining a reconstruction missing threshold value of the binocular camera according to the light bar missing information and the visual missing area, performing fusion vacancy filling on the three-dimensional binocular image based on the reconstruction missing threshold value to obtain a three-dimensional fusion image, and taking the three-dimensional fusion image as an analysis image of the brain surgery. And the image integrity of the brain surgery is improved.

A removably attachable phototherapy and vibration system for attachment to and treatment of skin beneath the eyes includes a therapeutic lamp platform for radiant lamps such as LEDs disposed in an assembly, and including a vibrating element, facing towards a patient and a plurality of light apertures substantially aligned with the LEDs for communicating lamp radiation from the lamps to a user. The lamps, vibrating element and associated circuitry are disposed so that the attachable surface is relatively smooth and seamless towards the patient. The system includes two devices one for the right and the other for the left eye of the user. Each device is attachable to beneath each eye with a suitable adhesive such as a reusable silicone strip. The system includes a charging dock for storing and charging the devices.

A method for determining a gaze convergence distance of a user of an eye movement tracking system. The method involves identifying a plurality of candidate convergence points along a combined gaze vector; and for each candidate convergence point, calculating a probability that the candidate convergence point is located at the gaze convergence distance based on: a) an angle between the left gaze vector and a straight line passing through the candidate convergence point from the left eye origin; b) an angle between the right gaze vector and a straight line passing through the candidate convergence point from the right eye origin; c) the distance between the left gaze vector and the candidate convergence point; and d) the distance between the right gaze vector and the candidate convergence point. The method then involves calculating a gaze convergence distance based on the candidate convergence point having the highest probability.

A determining device has a processor configured to determine that a driver has dozed off when degree of eye opening of a left eye and degree of eye opening of a right eye of the driver are at or below a reference degree of eye opening and an absolute value of difference between the degree of eye opening of the left eye and the degree of eye opening of the right eye does not exceed a reference difference or determine that the driver has not dozed off when the degree of eye opening of the left eye and the degree of eye opening of the right eye are at or below the reference degree and the absolute value of the difference between the degree of eye opening of the left eye and the degree of eye opening of the right eye exceeds a reference difference during a determining period.

The reference blurred image (Cp) is visually recognized by one of the two eyes (for example, the right eye), and the blurred images (Cp + 1, Cp + 2,...) are visually recognized by the other of the two eyes (for example, the left eye). Information relating to the sensitivity to the difference in the amount of blurring between blurred images (Cp + 1, Cp + 2,...) viewed by the other eye and a reference blurred image (Cp) viewed by one eye is acquired. Furthermore, the other eye is made to visually recognize the reference blurred image (Cp), the one eye is made to visually recognize the blurred images (Cp + 1, Cp + 2,...), and information relating to the sensitivity to the difference in the amount of blurring between the blurred images (Cp + 1, Cp + 2,...) visually recognized by one eye and the reference blurred image (Cp) visually recognized by the other eye is acquired.

The invention discloses an electronic sighting mark generating device for a synoptophore, relates to the field of optical systems, and is used for solving the problem that the types of sighting marks displayed in a traditional synoptophore are limited and are static. The device comprises a first display module and a second display module, the first display module and the second display module are arranged side by side, the left eye and the right eye of a testee face the first display module and the second display module respectively, the first display module is used for displaying a first test image and enabling light of the first test image to be emitted to the left eye, and the second display module is used for displaying a second test image. The second display module is used for displaying a second test image and enabling light of the second test image to be emitted to the right eye; and the sighting mark generation module is used for processing the image and generating image data, and respectively transmitting the image data to the first display module and the second display module, so that the first display module displays the first test image, and the second display module displays the second test image. Compared with the prior art that images are displayed through slides, more diversified test images can be displayed, and more dynamic test images can be displayed.

To provide a technique of displaying an aerial image in a distant view of a large screen.SOLUTION: In an embodiment of the present invention, the position of the point of view of an observer is between an aerial image which is presented by a presentation device for presenting an image and is formed in the air and the presentation device. The display device according to the embodiment of the present invention has a parallax compensation image generation unit for generating a presentation image to present by the presentation device by compensating the parallax of images for the right eye and for the left eye to display, on the basis of the distance from the position of the point of view and the aerial image and the parallax of an observer.SELECTED DRAWING: Figure 2

The invention relates to the field of image processing, and discloses a fixation point determination method, system and product. The method comprises the following steps: acquiring a first binocular image acquired by an eye camera aiming at the eyes of a user under a visible light condition; predicting a first left eye sight line of the user according to a left eye image in the first binocular image, and predicting a first right eye sight line of the user according to a right eye image in the first binocular image; predicting a first synthetic sight line of the user according to the first binocular image; the first synthetic sight line is a sight line associated with a scene camera; the scene camera is used for collecting a scene in a visual field range of the user; and determining a fixation point of the user based on two of the first left-eye sight line, the first right-eye sight line and the first synthetic sight line. By adopting the method, the near-to-eye tracking precision can be ensured under the visible light condition, and the deployment cost of eye movement tracking equipment can be reduced.

According to an embodiment of the disclosure, a media content rendering method may be provided. The method may include determining a plurality of eye attributes of a user, based on eye position for each of a left eye and a right eye. The method may include identifying eye movement for each of the left eye and right eye, based on the plurality of eye attributes. The method may include determining a constraint for a movement of each of the left eye and right eye, based on the eye movement. The method may include determining a binocular field-of-view (FOV) of the user, based on the constraint for each of the left eye and right eye. The method may include determining transformation values, based on the binocular FOV of the user. The method may include providing a transformed media content based on the transformation values to render the transformed media content.