Abstract: The present invention provides a graphics processing method and an eye-tracking system capable of providing adaptive foveated rendering. If the system latency of a current application is greater than the threshold latency under the existing rendering parameter, adjusted rendering parameters are provided by adjusting the existing rendering parameters according to the system latency of the current application. Foveated images may thus be generated according to the adjusted rendering parameters in order to maintain the balance of resolution quality and rendering performance.

Abstract: An optical system includes an eye-tracker and a head-mounted display for providing accurate eye-tracking in interactive virtual environment. The eye-tracker includes a sensor module captures one or multiple eye images of a user. The head-mounted display includes a processor and a display for presenting the user interface. The processor provides a user interface which includes one or multiple UI elements based on one or multiple gaze points of the user which are computed based on the one or multiple eye images, acquires the distance between an estimated gaze point of the user and each UI element, acquires the score of each UI element based on the distance between the estimated gaze point and each UI element, and sets a specific UI element with a highest score as the target UI element associated with the estimated gaze point of the user.

Abstract: A tensor accelerator includes two tile execution units and a bidirectional queue. Each of the tile execution units includes a buffer, a plurality of arithmetic logic units, a network, and a selector. The buffer includes a plurality of memory cells. The network is coupled to the plurality of memory cells. The selector is coupled to the network and the plurality of arithmetic logic units. The bidirectional queue is coupled between the selectors of the tile execution units.

Abstract: An optical system includes an eye-tracking module and a head-mounted display. The eye-tracking module includes a first sensor module configured to capture an eye image of a user. The head-mounted display includes a processor configured to provide a user interface based on a gaze point of the user computed based on the eye image, and a display configured to present the user interface. The user interface includes a virtual FoV and at least one UI element for receiving a gaze command from the user. The at least one UI element is arranged within the FoV of the user and located outside a range of the virtual FoV of the user interface.

Abstract: During an eye-tracking operation of an optical system, one or multiple eye images are captured. One or multiple gaze points of the user are computed based on the one or multiple eye images of the user. A user interface containing one or multiple UI elements is provided based on the one or multiple gaze points of the user. A performance map of the eye-tracking operation is provided based on the one or multiple eye images of the user. At least one UI element is adjusted based on the performance map of the eye-tracking operation.

Abstract: An optical system includes an eye-tracker and a head-mounted display for providing accurate eye-tracking in interactive virtual environment. The eye-tracker includes a sensor module captures one or multiple eye images of a user. The head-mounted display includes a processor and a display for presenting the user interface. The processor provides a user interface which includes one or multiple UI elements based on one or multiple gaze points of the user which are computed based on the one or multiple eye images, acquires the distance between an estimated gaze point of the user and each UI element, acquires the score of each UI element based on the distance between the estimated gaze point and each UI element, and sets a specific UI element with a highest score as the target UI element associated with the estimated gaze point of the user.

Abstract: A cognitive assessment system includes a computing device, a server coupled to the computing device and an eye tracking device coupled to the computing device. The computing device includes a cognitive assessment program with at least one cognitive assessment task for assessing at least one cognitive function of a user. The server includes a database. The database stores a historical task performance data of the user, a historical task performance model of the user, task performance data of a healthy population, and task performance data of a patient population. The eye tracking device is for capturing the eye movement data of the user.

Abstract: A tensor accelerator includes two tile execution units and a bidirectional queue. Each of the tile execution units includes a buffer, a plurality of arithmetic logic units, a network, and a selector. The buffer includes a plurality of memory cells. The network is coupled to the plurality of memory cells. The selector is coupled to the network and the plurality of arithmetic logic units. The bidirectional queue is coupled between the selectors of the tile execution units.

Abstract: The present invention provides a graphics processing method and an eye-tracking system capable of providing adaptive foveated rendering. If the system latency of a current application is greater than the threshold latency under the existing rendering parameter, adjusted rendering parameters are provided by adjusting the existing rendering parameters according to the system latency of the current application. Foveated images may thus be generated according to the adjusted rendering parameters in order to maintain the balance of resolution quality and rendering performance.

Abstract: A method of monitoring eye strain includes detecting the blink status, the vergence status and the pupil status of a user, and then determining whether the user encounters eye strain according to at least one of the blink status, the vergence status and the pupil status of the user. The method further includes facilitating the user to blink or informing the user of eye strain.

Abstract: An eye-tracking system includes a light-transmitting display module, a reflecting mirror, an image system, and a processing unit. The light-transmitting display module includes a first side and a second side. The imaging system is disposed on the second side of the light-transmitting display module and includes a camera lens and an image sensor. The camera lens is coated with an optical film for receiving the light reflected by a user face. The image sensor is configured to provide an eye image based on the light reflected by the user face. The processing unit is configured to analyze the eye image so as to acquire the facial characteristics associated with the eyes of the user, wherein the user face is located on the first side of the light-transmitting display module when the user puts on the eye-tracking system.

Abstract: A neural network system includes at least one memory and at least one processor. The memory is configured to store a front-end neural network, an encoding neural network, a decoding neural network and a back-end neural network. The processor is configured to execute the front-end neural network, the encoding neural network, the decoding neural network and the back-end neural network in the memory to perform operations including: utilizing the front-end neural network to output feature data; utilizing the encoding neural network to compress the feature data, and output compressed data which correspond to the feature data; utilizing the decoding neural network to decompress the compressed data, and output decompressed data which correspond to the feature data; and utilizing the back-end neural network to perform corresponding operations based on the decompressed data. A method of operating a neural network system is also disclosed herein.

Abstract: In a calibration process for an eye-tracking application, a calibration mark is displayed on an instrument, and the user is instructed to keep the gaze focused on the calibration mark. Next, a dynamic image is displayed on the instrument, and the user is instructed to move his head or the instrument as indicated by the dynamic image while keeping the gaze focused on the calibration mark. The ocular information of the user is recorded during the head movement or the instrument movement for calibrating the eye-tracking application.

Abstract: An eye-tracking module includes a power management module, a sensing module and a processor. The power management module is configured to supply power according to a power control signal for operating the eye-tracking module in a corresponding mode, and provide a power status signal associated its power supplying status. The sensing module includes at least one image sensor configured to capture an eye image of a user at a sampling rate. The processor is configured to acquire eye characteristics of the user, the eye-movement information of the user, or vision information including user gaze coordinates according to the eye image. The processor is also configured to adjust the sampling rate, adjust its output frequency and/or switch the operational mode of the eye-tracking module according to the power status signal.

Abstract: Eye-tracking methods and devices are provided for reducing operating workload. One sub frame of an eye frame and/or a scene frame is used to estimate the eyeball position. Such an approach can reduce operating workload and power consumption.

Abstract: Correction method and device for an eye-tracker are provided. A non-predetermined scene frame is provided and analyzed to obtain a salient feature information, which is in-turn used to correct an eye-tracking operation. The correction can be done at the initial or during the wearing of an eye-tracker.

Abstract: An adaptive eye-tracking calibration method includes generating an eye model acting as a current eye model; calibrating the eye model; comparing real-time pupil data set and pupil data set of the current eye model to obtain a pupil data difference when an event happens; and generating a new eye model if the pupil data difference is greater than a predetermined threshold.

Abstract: An image operation method and system for obtaining an eye's gazing direction are provided. The method and system employ multiple extraction stages for extracting eye-tracking features. An eye frame is divided into sub-frames, which are then sequentially temporarily stored in a storage unit. Launch features of sub frames are sequentially extracted from the sub frames by a first feature extraction stage, where a data of a former sub-frame is extracted before a data of a latter sub-frame is needed to be stored. Next, the remaining feature extraction stages apply a superposition operation on the launch features to obtain terminal features, which are then computed to obtain an eye's gazing direction.

Abstract: An eye-tracking module includes a power management module, a sensing module and a processor. The power management module is configured to supply power according to a power control signal for operating the eye-tracking module in a corresponding mode, and provide a power status signal associated its power supplying status. The sensing module includes at least one image sensor configured to capture an eye image of a user at a sampling rate. The processor is configured to acquire eye characteristics of the user, the eye-movement information of the user, or vision information including user gaze coordinates according to the eye image. The processor is also configured to adjust the sampling rate, adjust its output frequency and/or switch the operational mode of the eye-tracking module according to the power status signal.

Abstract: In a calibration process for an eye-tracking application, a calibration mark is displayed on an instrument, and the user is instructed to keep the gaze focused on the calibration mark. Next, a dynamic image is displayed on the instrument, and the user is instructed to move his head or the instrument as indicated by the dynamic image while keeping the gaze focused on the calibration mark. The ocular information of the user is recorded during the head movement or the instrument movement for calibrating the eye-tracking application.