Abstract: One embodiment is directed to a system for human-computer interface, comprising an input device configured to provide two or more dimensions of operational input to a processor based at least in part upon a rubbing contact pattern between two or more digits of the same human hand that is interpreted by the input device. The input device may be configured to provide two orthogonal dimensions of operational input pertinent to a three-dimensional virtual environment presented, at least in part, by the processor. The input device may be configured to detect the rubbing between a specific digit in a pen-like function against one or more other digits. The input device further may be configured to detect the rubbing between the specific digit in a pen-like function and one or more other digits in a receiving panel function.

Abstract: A user may interact and select positions in three-dimensional space arbitrarily through conversion of a two-dimensional positional input into a three-dimensional point in space. The system may allow a user to use one or more user input devices for pointing, annotating, or drawing on virtual objects, real objects or empty space in reference to the location of the three-dimensional point in space within an augmented reality or mixed reality session.

Abstract: A display subsystem for a virtual image generation system used by an end user, comprises first and second waveguide apparatuses, first and second projection subassemblies configured for introducing first and second light beams respectively into the first and second waveguide apparatuses, such that at least a first light ray and at least a second light ray respectively exit the first and second waveguide apparatuses to display first and second monocular images as a binocular image to the end user, and a light sensing assembly configured for detecting at least one parameter indicative of a mismatch between the displayed first and second monocular images as the binocular image.

Abstract: In some embodiments, eye tracking is used on an AR or VR display system to determine if a user of the display system is blinking or otherwise cannot see. In response, current drain or power usage of a display associated with the display system may be reduced, for example, by dimming or turning off a light source associated with the display, or by configuring a graphics driver to skip a designated number of frames or reduce a refresh rate for a designated period of time.

Abstract: Examples of the disclosure describe systems and methods for estimating acoustic properties of an environment. In an example method, a first audio signal is received via a microphone of a wearable head device. An envelope of the first audio signal is determined, and a first reverberation time is estimated based on the envelope of the first audio signal. A difference between the first reverberation time and a second reverberation time is determined. A change in the environment is determined based on the difference between the first reverberation time and the second reverberation time. A second audio signal is presented via a speaker of a wearable head device, wherein the second audio signal is based on the second reverberation time.

Abstract: Systems and methods for eye image segmentation and image quality estimation are disclosed. In one aspect, after receiving an eye image, a device such as an augmented reality device can process the eye image using a convolutional neural network with a merged architecture to generate both a segmented eye image and a quality estimation of the eye image. The segmented eye image can include a background region, a sclera region, an iris region, or a pupil region. In another aspect, a convolutional neural network with a merged architecture can be trained for eye image segmentation and image quality estimation. In yet another aspect, the device can use the segmented eye image to determine eye contours such as a pupil contour and an iris contour. The device can use the eye contours to create a polar image of the iris region for computing an iris code or biometric authentication.

Abstract: Examples of wearable devices that can present to a user of the display device an audible or visual representation of an audio file comprising a plurality of stem tracks that represent different audio content of the audio file are described. Systems and methods are described that determine the pose of the user; generate, based on the pose of the user, an audio mix of at least one of the plurality of stem tracks of the audio file; generate, based on the pose of the user and the audio mix, a visualization of the audio mix; communicate an audio signal representative of the audio mix to the speaker; and communicate a visual signal representative of the visualization of the audio mix to the display.

Abstract: Disclosed herein are systems and methods for presenting mixed reality audio. In an example method, audio is presented to a user of a wearable head device. A first position of the user's head at a first time is determined based on one or more sensors of the wearable head device. A second position of the user's head at a second time later than the first time is determined based on the one or more sensors. An audio signal is determined based on a difference between the first position and the second position. The audio signal is presented to the user via a speaker of the wearable head device. Determining the audio signal comprises determining an origin of the audio signal in a virtual environment. Presenting the audio signal to the user comprises presenting the audio signal as if originating from the determined origin. Determining the origin of the audio signal comprises applying an offset to a position of the user's head.

Abstract: A wearable augmented reality head-mounted display system can be configured to pass light from the world forward a wearer wearing the head-mounted system into an eye of the wearer. The head-mounted display system can include an optical display that is configured to output light to form an image. The system may include one or more waveguides that are disposed to receiving the light from the display. A variable power reflector can be disposed on the forward side of the one or more waveguides. The reflector can be configured to have an optical power that is adjustable upon application of an electrical signal.

Abstract: Systems and methods for rendering audio signals are disclosed. In some embodiments, a method may receive an input signal including a first portion and the second portion. A first processing stage comprising a first filter is applied to the first portion to generate a first filtered signal. A second processing stage comprising a second filter is applied to the first portion to generate a second filtered signal. A third processing stage comprising a third filter is applied to the second portion to generate a third filtered signal. A fourth processing stage comprising a fourth filter is applied to the second portion to generate a fourth filtered signal. A first output signal is determined based on a sum of the first filtered signal and the third filtered signal. A second output signal is determined based on a sum of the second filtered signal and the fourth filtered signal.

Abstract: Systems and methods for estimating a gaze vector of an eye using a trained neural network. An input image of the eye may be received from a camera. The input image may be provided to the neural network. Network output data may be generated using the neural network. The network output data may include two-dimensional (2D) pupil data, eye segmentation data, and/or cornea center data. The gaze vector may be computed based on the network output data. The neural network may be previously trained by providing a training input image to the neural network, generating training network output data, receiving ground-truth (GT) data, computing error data based on a difference between the training network output data and the GT data, and modifying the neural network based on the error data.

Abstract: Systems and methods for estimating a layout of a room are disclosed. The room layout can comprise the location of a floor, one or more walls, and a ceiling. In one aspect, a neural network can analyze an image of a portion of a room to determine the room layout. The neural network can comprise a convolutional neural network having an encoder sub-network, a decoder sub-network, and a side sub-network. The neural network can determine a three-dimensional room layout using two-dimensional ordered keypoints associated with a room type. The room layout can be used in applications such as augmented or mixed reality, robotics, autonomous indoor navigation, etc.

Abstract: A system and method for operating a depth sensor. A configuration operation can be performed by storing a first sequence of operation steps which define a first depth sensing mode of operation, and a second sequence of operation steps which define a second depth sensing mode of operation, in the memory. In response to a first request for depth measurements according to the first depth sensing mode of operation, the depth sensor can be operated in the first mode of operation by causing it to execute the first sequence of operation steps. In response to a second request for depth measurements according to the second depth sensing mode of operation, and without performing an additional configuration operation, the depth sensor can be operated in the second mode of operation by causing it to execute the second sequence of operation steps.

Abstract: A method of operating an eyepiece waveguide having a first region and a second region includes directing light from a first projector to impinge on a first incoupling grating (ICG) and light from a second projector to impinge on a second ICG. Light from the first projector is diffracted into a first portion and a second portion of the first region of the eyepiece waveguide and out of the eyepiece waveguide. Light from the second projector is diffracted into a first portion and a second portion of the second region of the eyepiece waveguide and out of the eyepiece waveguide.

Abstract: Examples of the disclosure describe systems and methods for presenting an audio signal to a user of a wearable head device. According to an example method, a source location corresponding to the audio signal is identified. For each of the respective left and right ear of the user, a virtual speaker position, of a virtual speaker array, is determined, the virtual speaker position collinear with the source location and with a position of the respective ear. For each of the respective left and right ear of the user, a head-related transfer function (HRTF) corresponding to the virtual speaker position and to the respective ear is determined; and the output audio signal is presented to the respective ear of the user via one or more speakers associated with the wearable head device. Processing the audio signal includes applying the HRTF to the audio signal.

Abstract: A neural network in multi-task deep learning paradigm for machine vision includes an encoder that further includes a first, a second, and a third tier. The first tier comprises a first-tier unit having one or more first-unit blocks. The second tier receives a first-tier output from the first tier at one or more second-tier units in the second tier, a second-tier unit comprises one or more second-tier blocks, the third tier receives a second-tier output from the second tier at one or more third-tier units in the third tier, and a third-tier block comprises one or more third-tier blocks. The neural network further comprises a decoder operatively the encoder to receive an encoder output from the encoder as well as one or more loss function layers that are configured to backpropagate one or more losses for training at least the encoder of the neural network in a deep learning paradigm.

Abstract: A computer implemented method for recognizing a hand gesture using a random forest model includes training the random forest model. The method also includes obtaining image data. The method further includes clustering a plurality of pixels from the image data to generate a plurality of clusters. Moreover, the method includes analyzing the plurality of clusters using a rejection cascade to generate a plurality of selected candidates. In addition, the method includes analyzing the plurality of selected candidates using a classification decision tree from the random forest model. The method also includes skeletonizing the plurality of selected candidates to generate a one dimension plus branches hand model. The method further includes analyzing the one dimension plus branches hand model using a regression decision tree from the random forest model.

Abstract: A display assembly displays a virtual object in a select location wherein an eye viewing the virtual object has an expected gaze direction. Deformation of the display assembly is detected. The deformation causes the virtual object to be viewable in an altered location wherein the eye has altered gaze direction. The virtual object may be displayed in a corrected location wherein the eye viewing the virtual object in the correct location has a corrected gaze direction that is moved closer to the expected gaze direction than the altered gaze direction.

Abstract: This disclosure relates to the use of variable-pitch light-emitting devices for display applications, including for displays in augmented reality, virtual reality, and mixed reality environments. In particular, it relates to small (e.g., micron-size) light emitting devices (e.g., micro-LEDs) of variable pitch to provide the advantages, e.g., of compactness, manufacturability, color rendition, as well as computational and power savings. Systems and methods for emitting multiple lights by multiple panels where a pitch of one panel is different than pitch(es) of other panels are disclosed. Each panel may comprise a respective array of light emitters. The multiple lights may be combined by a combiner.

Abstract: Systems and methods of presenting an output audio signal to a listener located at a first location in a virtual environment are disclosed. According to embodiments of a method, an input audio signal is received. A first intermediate audio signal corresponding to the input audio signal is determined, based on a location of the sound source in the virtual environment, and the first intermediate audio signal is associated with a first bus. A second intermediate audio signal is determined. The second intermediate audio signal corresponds to a reverberation of the input audio signal in the virtual environment. The second intermediate audio signal is determined based on a location of the sound source, and further based on an acoustic property of the virtual environment. The second intermediate audio signal is associated with a second bus. The output audio signal is presented to the listener via the first and second buses.