An electronic apparatus includes a memory which temporarily stores image data of an image captured by an imaging device, and a processor which processes image data stored in the memory. The processor processes image data of plural images captured by the imaging device at predetermined time intervals and stored in the memory, detects face areas with faces captured therein from among the plural images based on first-resolution image data and second-resolution image data, and determines whether or not the face areas are consecutively detected from the plural images. Further, when determining that the state is changed between a state where face areas are consecutively detected and a state where face areas are not consecutively detected while performing processing to detect the face areas based on first-resolution image data, the processor detects face areas from among the plural images based on the second-resolution image data.

Methods and apparatus relating to techniques for provision of low power foveated rendering to save power on GPU (Graphics Processing Unit) and/or display are described. In various embodiment, brightness/contrast, color intensity, and/or compression ratio applied to pixels in a fovea region are different than those applied in regions surrounding the fovea region. Other embodiments are also disclosed and claimed.

This application discloses a method for adjusting image luminance performed at an electronic device. The method includes: determining a target pixel with original luminance lower than a luminance threshold in an original image, the luminance threshold being determined according to luminance of pixels in the original image; obtaining a luminance distribution intensity of pixels adjacent to the target pixel; determining a difference between the luminance threshold and the luminance distribution intensity of the adjacent pixels; and adjusting the target pixel to corresponding target luminance according to the difference and the original luminance of the target pixel. According to this application, a changing characteristic of relative luminance between the target pixel and the adjacent pixel is reserved. The luminance adjustment is more consistent with the luminance propagation of the original image, thereby achieving a technical effect of improving an adjustment effect of luminance adjustment on an image.

A method comprising the steps of obtaining at least one frame of input data from at least one sensor, the frame of input data representative of a real-world environment at a given time. The frame is analysed to determine at least a foveal region within the frame, and at least a method for generating depth information associated with the real-world environment based on the frame of input data is selected. The method is applied to the foveal region to generate depth information associated with the foveal region, and at least the depth information associated with the foveal region is outputted.

In various examples, apparatuses, systems, and techniques to perform offline image signal processing of source image data to generate target image data. In at least one embodiment, data collection using exposure and calibration setting of an image sensor is performed to generate source image data, which is then processed by using offline image signal processing to generate target data.

A volumetric data structure models a particular volume representing the particular volume at a plurality of levels of detail. A first entry in the volumetric data structure includes a first set of bits representing voxels at a first level of detail, the first level of detail includes the lowest level of detail in the volumetric data structure, values of the first set of bits indicate whether a corresponding one of the voxels is at least partially occupied by respective geometry, where the volumetric data structure further includes a number of second entries representing voxels at a second level of detail higher than the first level of detail, the voxels at the second level of detail represent subvolumes of volumes represented by voxels at the first level of detail, and the number of second entries corresponds to a number of bits in the first set of bits with values indicating that a corresponding voxel volume is occupied.

Techniques to improve performance of matrix multiply operations are described in which a compute kernel can specify one or more element-wise operations to perform on output of the compute kernel before the output is transferred to higher levels of a processor memory hierarchy.

According to example embodiments, an Image View Aggregator identifies a frontal view of an item within an image. The Image View Aggregator identifies at least one reflection view of the item within the image. Each reflection view of the item having been captured off a corresponding reflective physical surface. The Image View Aggregator extracts the frontal view of the item and each reflection view of the item from the image. The Image View Aggregator generates a representation of the item based at least on the extracted frontal view of the item and each extracted reflection view of the item.

Implementations generally relate to providing augmented reality in a web browser. In one implementation, a method includes capturing images of a physical scene with a camera of a device. The method further includes determining motion of the camera using six degrees of freedom (6DoF) markerless tracking. The method further includes overlaying virtual three-dimensional (3D) content onto a depicted physical scene in the images, resulting in augmented reality (AR) images. The method further includes rendering the AR images in a browser of the device.

A rendering processor assigns varying logical pixel dimensions to regions of an image frame and rendering pixels of the image frame based on the logical pixel dimensions. The rendering processor renders in highest resolution (i.e., with smaller logical pixel dimensions) those areas of the image that are more important (on which the viewer is expected to focus (the “foveal region”), or regions with little-to-no motion), and renders in lower resolution (i.e., with larger logical pixel dimensions) those areas of the image outside the region of interest, or regions that are speedily moving, so that loss of detail in those regions will be less noticeable to the viewer. For regions with less detail or greater magnitude of motion, larger logical pixel dimensions reduce the computational workload without affecting the quality of the displayed graphics as perceived by a user.