Given an input image, an image enhancement task, and no external examples available to train on, an Image-Specific Deep Network is constructed tailored to solve the task for this specific image. Since there are no external examples available to train on, the network is trained on examples extracted directly from the input image itself. The current solution solves the problem of Super-Resolution (SR), whereas the framework is more general and is not restricted to SR.

Techniques for synthesizing intermediary frames for long exposure images are disclosed. Video stream data, comprising a plurality of arriving frames, is received. An indication that the video stream data should be processed into a long exposure image that incorporates one or more synthesized frames is received. A set of arriving frames is used to generate at least one synthesized frame. At least one received frame is blended with the at least one generated synthesized frame to form a long exposure image. The long exposure image is provided as output.

A system, method, and apparatus for generating a normalization of a single two-dimensional image of an unconstrained human face. The system receives the single two-dimensional image of the unconstrained human face, generates an undistorted face based on the unconstrained human face by removing perspective distortion from the unconstrained human face via a perspective undistortion network, generates an evenly lit face based on the undistorted face by normalizing lighting of the undistorted face via a lighting translation network, and generates a frontalized and neutralized expression face based on the evenly lit face via an expression neutralization network.

A method and a system for processing an image and transform it into a high resolution and high-definition image using a computationally efficient image transformation procedure is provided. The transformation of the image comprises first transforming the image, also referred to as an intensity image, into a layered distance field (DF) image comprising an ordered sequence of multiple layers. Each layer in the ordered sequence is associated with a DF procedure and a set of rules for mapping the DF values to intensity values of the respective layer. The result of applying the DF procedures to each location in the intensity image is a transformed intensity image, which is of high definition and high resolution. The application of the DF procedures is governed by a stopping criteria based on error values between the intensity image and a reconstructed intensity image.

The present invention relates to the technical field of producing artificial contrast-enhanced radiological images by way of machine learning methods.

A method for providing a prediction of a representation of an examination region that was generated using a medical image technique involving a contrast agent may include receiving a first representation in frequency space of an examination region of an examination object, receiving a second representation in the frequency space of the examination region of the examination object, providing the first representation and the second representation as an input to a predictive machine learning model that is configured to provide, as an output, a prediction of a representation in the frequency space of the examination region with an amount of the contrast agent administered during a medical imaging technique, receiving the output of the predictive machine learning model based on the input, and converting the output of the predictive machine learning model to a representation in real space of the examination region of the examination object.

An electronic apparatus includes: an artificial intelligence AI processor, configured to select a first image processing model from a plurality of image processing models based on scenario information. The AI processor performs first image signal processing on a first image signal by using the first image processing model, to obtain a second image signal. The first image signal is obtained based on first image data output by an image sensor. The scenario information represents feature classification of the first image signal. An image signal processor ISP, configured to perform second image signal processing on the second image signal, obtains a first image processing result. The electronic apparatus provided in embodiments of this application can improve an image processing effect.

The present disclosure provides an image processing method and an image processing apparatus. The image processing method includes: obtaining a to-be-converted SDR image; using a first convolutional network to perform feature analysis on the SDR image, to obtain N weights of the SDR image; where the N weights are respectively configured to characterize proportions of color information of the SDR image to color information characterized in preset N 3D lookup tables, the N 3D lookup tables are configured to characterize color information of different types; obtaining a first 3D lookup table for the SDR image according to the N weights and the N 3D lookup tables; using the first 3D lookup table to adjust the color information of the SDR image to obtain an HDR image; and using a second convolutional neural network to perform refinement correction on the HDR image to obtain an output image.

A control device 20 includes an image acquisition unit 203 configured to acquire a radiographic image obtained by irradiating a subject F with radiation and capturing an image of the radiation passing through the subject F, a noise map generation unit 204 configured to derive an evaluation value obtained by evaluating spread of a noise value from a pixel value of each pixel in the radiographic image on the basis of relationship data indicating a relationship between the pixel value and the evaluation value and generate a noise map that is data in which the derived evaluation value is associated with each pixel in the radiographic image, and a processing unit 205 configured to input the radiographic image and the noise map to a trained model 207 constructed in advance through machine learning and execute image processing of removing noise from the radiographic image.

A processor-implemented method includes: obtaining a plurality of image frames acquired for a scene within a predetermined time; determining loss values respectively corresponding to the plurality of image frames; determining a reference frame among the plurality of image frames based on the loss values; and generating a final image of the scene based on the reference frame.