7 results about "Image plane" patented technology

An optical amplification method which includes a spectral filtering method for selecting one of several pre-recorded spectral filter patterns and applies this filter to the spectral distribution of a multi-wavelength optical signal transmitted from an input to an output fiber. Optical filtering may be accomplished using free-space bulk optical diffractive wavelength demultiplexing, combined with a fixed (permanent) spatial pattern located in the spectrally dispersed image plane, to transfer to the optical output fiber the input signal with its spectrum selectively attenuated.

The present disclosure provides embodiments of improved systems, head mounted devices (HMDs) and methods for adjusting a position of an HMD with respect to a user based on one or more measurements, which are specific to the user. One embodiment of the improved method includes receiving data from an eye tracking system of an HMD worn by a user, where the data is generated by the eye tracking system in response to the user gazing upon a virtual image plane of the HMD; using the received data to calculate a distance between the virtual image plane of the HMD and a pupil plane of the user; and adjusting a position of the HMD with respect to the user based on the calculated distance.

Embodiments of the invention provide a virtual distortion image generation method and device, and relates to the technical field of virtual reality. The method comprises the following steps: determining a to-be-rendered color parameter of each pixel point in an image plane of a virtual camera, wherein the to-be-rendered color parameter is obtained by mapping each virtual object in a three-dimensional simulation scene into an image plane under the condition that camera distortion does not exist; determining a distortion pixel point corresponding to each pixel point in the image plane based on a distortion parameter corresponding to a camera distortion type; and rendering the color parameter of each pixel point into the to-be-rendered color parameter of the distorted pixel point corresponding to each pixel point to obtain the virtual distortion image. Compared with the prior art, by applying the scheme provided by the embodiment of the invention, pixel loss in the virtual distortion image can be avoided, and image quality is improved.

The invention discloses an AR-HUD/LCD omnidirectional target detection and display method based on human eye attention, and belongs to the technical field of automatic driving, automobile environment perception, and intelligent cabins. The method comprises the following steps: acquiring a face image of a driver through a monocular camera; calibrating the monocular camera to obtain an internal reference matrix and a distortion parameter of the camera; carrying out distortion removal and graying on the image to obtain a self-quotient image; extracting a face area through a face detector, and then extracting coordinates of left and right pupils in an image plane of the monocular camera; according to the pupil image plane coordinates and pupil distance constraints, solving driver eye position space coordinates used for AR-HUD/LCD virtual image display. A neural network is used for detecting changes of a face angle of a driver, a driver attention vector is obtained, a target is detected in a vehicle-mounted panoramic image or an omnidirectional radar system according to the attention vector, and a detection object in an attention vector area is displayed through AR-HUD/LCD. According to the invention, system computing resources are saved, and the real-time performance of the detection system is improved.