Techniques for generating a high resolution full color output image from lower resolution sparse color input images are disclosed. A camera generates images. The camera's sensor has a sparse Bayer pattern. While the camera is generating the images, IMU data for each image is acquired. The IMU data indicates a corresponding pose the camera was in while the camera generated each image. The images and IMU data are fed into a motion model, which performs temporal filtering on the images and uses the IMU data to generate a red-only image, a green-only image, a blue-only image, and a monochrome image. The color images are up-sampled to match the resolution of the monochrome image. A high resolution output color image is generated by combining the up-sampled images and the monochrome image.

In one embodiment, a computing system may determine, a predicted eye position of a viewer corresponding to a future time moment for displaying a frame. The system may generate a first correction map for the frame based on the predicted eye position of the viewer. The system may retrieve one or more second correction maps used for correcting one or more proceeding frames. The system may generate a third correction map based on the first correction map generated based on the predicted eye position of the viewer and the one or more second correction maps used for correcting the one or more proceeding frames. The system may adjust pixel values of the frame based at least on the third correction map. The system may output the frame with the adjusted pixel values to a display.

An apparatus includes a memory and a processor. The memory may be configured to store image data of an input image. The processor may be configured to detect one or more objects in the input image using a quantized multi-stage object detection network, where quantization of the quantized multi-stage object detection network includes (i) generating quantized image data by performing a first data range analysis on the image data of the input image, (ii) generating a feature map and proposal bounding boxes by applying a region proposal network (RPN) to the quantized image data, (iii) performing a region of interest pooling operation on the feature map and a plurality of ground truth boxes corresponding to the proposal bounding boxes generated by the RPN, (iv) generating quantized region of interest pooling results by performing a second data range analysis on results from the region of interest pooling operation, and (v) applying a region-based convolutional neural network (RCNN) to the quantized region of interest pooling results.

Techniques are described for efficient high-resolution output of an image captured using a high-pixel-count image sensor based on pixel binning followed by luminance-guided umsampling. For example, an image sensor array is configured according to a red-green-blue-luminance (RGBL) CFA pattern, such that at least 50-percent of the imaging pixels of the array are luminance (L) pixels. Pixel binning is used during readout of the array to concurrently generate a downsampled RGB capture frame and a downsampled L capture frame. Following the readout, the L capture frame is upsampled (e.g., by upscaling and interpolation) to generate an L guide frame with 100-percent luminance density. An upsampled RGB frame can then be generated by interpolating the RGB capture frame based both on known neighboring RGB information (e.g., from the RGB capture frame and previously interpolated information), as adjusted based on local luminance information from the L guide frame.

A system for automated comprehensive assessment of clinical lumbar MRIs includes a MRI standardization component that reads MRI data from raw lumbar MRI files, uses an artificial intelligence (AI) model to convert the raw MRI data into a standardized format. A core assessment component automatically generates MRI assessment results, including multi-tissue anatomical annotation, multi-pathology detection and multi-pathology progression prediction based on the structured MRI data package. The core assessment component contains a semantic segmentation module that utilizes a deep learning artificial intelligence (AI) model to generate an MRI assessment results that contains multi-tissue anatomical annotation, a pathology detection module to generate multi-pathology detection, and a pathology progression prediction module to generate multi-pathology progression prediction. A model optimization component archives clinical MRI data and MRI assessment results based on comments provided by a specialist, and periodically optimizes the AI deep learning model of the core assessment component.

Users often desire to capture certain images from an application. Existing methods of capturing images can result in low-resolution images due to limitations of the display device providing the images. This disclosure provides a method of capturing higher resolution images from source images. Techniques are also disclosed to reduce the storage size associated with the higher resolution images. Through capturing low-resolution versions of the same source images, image effects can be captured and applied to the higher resolution images where those image effects may be altered or missing. Frequency spectrum combination can be used to combine the low-resolution image data and the higher resolution image data. The higher resolution images can be processed using a segmentation scheme, such as tiling, without reducing or limiting the image effects.

The present invention provides image processing apparatus, an image processing system and an image processing method, whereby the accuracy of evaluation can be improved. Image processing apparatus for correcting a captured image includes an image acquisition unit that acquires block pixels, an image conversion unit that converts the lightness/darkness, density, luminance and color space of the image based on the RGB values contained in the block pixels, and an output unit that outputs the converted image, and the image conversion unit further includes a binarization processing unit that binarizes the image, calculates the area ration of lightness and darkness in the image, and specifies the light and dark fields, a density conversion unit that performs conversion into wavelengths corresponding to the RGB values, a luminance conversion unit that performs conversion into luminance of the maximum wavelength that is visible to human eye, and a color-space conversion unit that performs conversion into HSV values representing a color space of color tones.

Disclosed is a method for generating high resolution point cloud data for electro-anatomical mapping comprising receiving sparsely measured point cloud data having a plurality of data points. Surface mesh data comprising mesh points defining triangles on a myocardial surface is generated. The point cloud data is mapped to the surface mesh data. For each point of the surface mesh data that cannot be mapped to the point cloud data because there is a missing data point in point cloud data, an interpolation operation is performed based on the point cloud data within the neighbourhood of the point to generate a value for the missing data point. The interpolation operation is repeated N times. For every repetition, a difference between the value for the missing data point generated from the current iteration and the value for the missing data point generated from the immediately preceding iteration is compared, until the difference is below a threshold.

An electronic device may include an electronic display to display an image based on processed image data. The electronic device may also include image processing circuitry to determine a hierarchical grid having multiple grid points divided into grid partitions. A first set of grid points associated with a first set of grid partitions may include a first set of mappings to corresponding coordinates of input image data in a source frame. The image processing circuitry may also interpolate between the first set of grid points to determine a second set of grid points of having a second set of mappings to corresponding coordinates of the input image data based on the first set of mappings. The image processing circuitry may also generate the processed image data by applying the first set of mappings and the second set of mappings to the input image data.

The disclosure extends to methods, systems, and computer program products for producing an image in light deficient environments and associated structures, methods and features. The features of the systems and methods described herein may include providing improved resolution and color reproduction.