Methods and apparati for inserting face and hair information from a source video (401) into a destination (driver) video (402) while mimicking pose, illumination, and hair motion of the destination video (402). An apparatus embodiment comprises an identity encoder (404) configured to encode face and hair information of the source video (401) and to produce as an output an identity vector; a pose encoder (405) configured to encode pose information of the destination video (402) and to produce as an output a pose vector; an illumination encoder (406) configured to encode head and hair illumination of the destination video (402) and to produce as an output an illumination vector; and a hair motion encoder (414) configured to encode hair motion of the destination video (402) and to produce as an output a hair motion vector. The identity vector, pose vector, illumination vector, and hair motion vector are fed as inputs to a neural network generator (410). The neural network generator (410) is configured to generate, in response to the four inputs, a composite video (403) comprising face and hair information from the source video (401) inserted into the destination video (402).

The method includes generating, for each of a plurality of original images, a first artificially degraded image by applying a first image-artifact-generation logic on each of the original images; and generating the program logic by training an untrained version of a first machine-learning logic that encodes a first artifacts-removal logic on the original images and their respectively generated first degraded images; and returning the trained first machine-learning logic as the program logic or as a component thereof. The first image-artifact-generation logic is A) an image-acquisition-system-specific image-artifact-generation logic or B) a tissue-staining-artifact-generation logic.

A method and device are provided for inpainting image. An electronic device can be used for: determining an inpainting region of the image, wherein the inpainting region includes a defective region; obtaining original texture information of the inpainting region; determining inpainting pixel blocks and backup pixel blocks, wherein the inpainting pixel blocks include inpainting pixels and first type pixels; inpainting all inpainting pixels of the inpainting pixel blocks based on the backup pixel blocks; and superimposing the original texture information in the inpainting region.

An image processing apparatus that operates in a plurality of modes including a first mode for performing high image quality processing of image data and a second mode for performing high-speed processing of image data, performs the following processing. More specifically, the apparatus performs, in accordance with the mode to be operated, the settings of each of an operation to output image data from a first DMAC (Direct Memory Access Controller) to a common bus, an operation to input the image data from the common bus to an image processing unit, an operation of the image processing unit, an operation to output the image data from the image processing unit to the common bus, and an operation to output the image data from the common bus to a second DMAC.

An image processing method, an image processing device, an electronic device, and a storage medium are provided. The method comprises: performing image segmentation on a first image which is to be processed, obtaining an initial mask image according to an image segmentation result, in response to determining that the first image satisfies a preset sky area replacement condition according to the initial mask image, obtaining a target mask image by performing guided filtering on the initial mask image by using a greyscale image of the first image as a guide image, acquiring a target sky scene, and obtaining a second image by performing replacement on the sky area in first the image according to the target mask image and the target sky scene.

Techniques are described for inpainting of image data with a missing region. In an embodiment, at each iteration, the process determines a corresponding missing boundary region of the missing region and generates a collection of boundary patches for the missing boundary region. Based on comparing a boundary patch from the collection to source patches from a known source region of image data, the process generates replacement patches for the missing boundary region. When a boundary pixel data unit corresponds to multiple replacement pixel data units from different replacement patches, the process aggregates the multiple replacement pixel data units to generate an updated boundary pixel data unit. In an embodiment, the process performs convolution using the updated and previously known region of the image data.

A method for automatically enhancing an image from a device includes obtaining a first image using an imaging device. Recognition software is configured to recognize an object or individual in the first image. An initial image profile is configured based on the first image. Editing software is used to edit at least one attribute of the initial image profile. At least one subsequent image is taken or received. The recognition software is used to recognize the object or individual in the at least one subsequent image. The at least one attribute of the at least one subsequent image is automatically edited based on the initial image profile.

A method for segmenting a power line image in real time based on self-supervised learning includes: inputting an input power line sample image and power line sample image mask set for the same batch of images into a region growing algorithm to obtain a single power line sub-image and single power line mask set; randomly extracting at least one single power line image pair for combination, and combining the single power line image pair with a random background picture to generate a power line random background fusion image and power line random background mask set; and carrying out random non-repetitive region growing to obtain image inpainting regions, forming a segmentation mask with the image inpainting regions, obtaining power line segmentation images through an image inpainting algorithm, inputting the power line segmentation images into a power line real-time segmentation network for training, and carrying out predicted segmentation.

An image processing apparatus, including a memory configured to store one or more instructions; and at least one processor configured to execute the one or more instructions to: based on a first image and a probability model, optimize an estimated pixel value and estimated gradient values of each pixel of an original image corresponding to the first image, obtain an estimated original image based on the optimized estimated pixel value of the each pixel of the original image, obtain a decontour map based on the optimized estimated pixel value and the estimated gradient values of the each pixel of the original image, and generate a second image by combining the first image with the estimated original image based on the decontour map.

Systems and methods for specular reflection reduction in endoscope visualizations are described. A method includes receiving an image including a region of specular reflection. The method includes detecting the region of specular reflection in the image. The method includes estimating image information for a portion of the region of specular reflection. The method also includes reconstructing the image including the image information populated into the region of specular reflection.