The present invention discloses a multi-frame image super resolution system that utilizes both deep learning models and traditional models of enhancing the resolution of an image so that minimal computational resources are used. A frame alignment module of the invention aligns the frames of the image after which a processing module configured within the system process the Y and the UV channels of the image by using multiple deep and traditional resolution enhancement models. A merging unit merges the output of the processors to produce a super resolution image incorporating the advantages of both of the image enhancement methods.

Embodiments of the present disclosure provide a method, an electronic device, and a program product for image processing. In one embodiment, a method may include: at an edge node of a network, obtaining a first image generated based on data associated with a target event, wherein the first image has a first resolution ratio. Additionally, the method may further include: sending a second image converted from the first image to a terminal device, wherein the second image has a second resolution ratio higher than the first resolution ratio. According to the embodiments of the present disclosure, by rendering a low-resolution-ratio image at a cloud server and transmitting the image to an edge node or a terminal device for reconstructing a high-resolution-ratio image, the bandwidth and time delay of high-definition image transmission can be significantly reduced, so that the user experience is improved.

An imaging device, according to one embodiment of the present invention, comprises: an input unit for receiving first Bayer data having a first resolution and a noise level; and a convolutional neural network for outputting second Bayer data having a second resolution by using the noise level and the first Bayer data.

An image capturing circuit that can be applied to an image capturing apparatus, including a pixel array having photoelectric conversion elements arranged in a matrix, a row selection circuit capable of reading out still image data with a first resolution and LV moving image data with a second resolution lower than the first resolution from the pixel array, an image memory for storing the read still image data and the read LV moving image data, and a reduction circuit that converts the still image data to reduced image data with a third resolution lower than the first resolution. The still image data is read out from the memory and output after converted to the reduced still image data, and the still image data is also output without being converted. The LV moving image data is read out from the memory and output without being converted.

Methods and systems are provided for using temporal supersampling to increase a displayed resolution associated with peripheral region of a foveated rendering view. A method for enabling reconstitution of higher resolution pixels from a low resolution sampling region for fragment data is provided. The method includes an operation for receiving a fragment from a rasterizer of a GPU and for applying temporal supersampling to the fragment with the low resolution sampling region over a plurality of prior frames to obtain a plurality of color values. The method further includes an operation for reconstituting a plurality of high resolution pixels in a buffer that is based on the plurality of color values obtained via the temporal supersampling. Moreover, the method includes an operation for sending the plurality of high resolution pixels for display.

An image processing method and a device, where the image processing method is performed by a terminal having a digital zoom function, and the method includes determining a target zoom magnification based on a selection input of a user, collecting a to-be-processed image, and processing the to-be-processed image using a target super-resolution convolutional neural network model to obtain a processed image corresponding to the target zoom magnification, where the target super-resolution convolutional neural network model is obtained by training a super-resolution convolutional neural network model using a high-definition training image, a low-definition training image, and a mask image.

The present disclosure relates to image processing. The method includes acquiring at least one of a backward propagation feature of an (x+1)th video frame in a video segment or a forward propagation feature of an (x−1)th video frame in the video segment. The video segment includes N video frames, N being an integer greater than 2, and x being an integer. The method further includes deriving a reconstruction feature of the xth video frame from at least one of the xth video frame, the backward propagation feature of the (x+1)th video frame, or the forward propagation feature of the (x−1)th video frame, and deriving a target video frame corresponding to the xth video frame by reconstructing the xth video frame based on the reconstruction feature of the xth video frame. The target video frame has resolution higher than that of the xth video frame.

A signal processing device and an image display apparatus including the same are disclosed. The signal processing device includes a scaler configured to scale input images of various resolutions to a first resolution, a resolution enhancement processor configured to perform learning on the input images and to output a residual image of the first resolution, and an image output interface configured to output an output image of the first resolution based on a scaling image from the scaler and the residual image from the resolution enhancement processor, and the image output interface changes a weight and an application strength of the residual image according to the area of the input image.

This disclosure relates generally to a method and system for multi-modal image super-resolution. Conventional methods for multi-modal image super-resolution are performed using joint image based filtering, deep learning and dictionary based approaches which require large datasets for training. Embodiments of the present disclosure provide a joint optimization based transform learning framework wherein a high-resolution (HR) image of target modality is reconstructed from a HR image of guidance modality and a low-resolution (LR) image of target modality. A set of parameters, transforms, coefficients and weight matrices are learnt jointly from a training data which includes a HR image of guidance modality, a LR image of target modality and a HR image of target modality. The learnt set of parameters are used for reconstructing a HR image of target modality. The disclosed joint optimization transform learning framework is used in remote sensing, environment monitoring and so on.

Methods and systems for high-resolution image manipulation are disclosed. An original high-resolution image to be manipulated is obtained, as well as a driving signal indicating a manipulation result. The original high-resolution image is down-sampled to obtain a low-resolution image to be manipulated. Using a trained manipulation generator, a low-resolution manipulated image and a motion field are generated from the low-resolution image. The motion field represent pixel displacements of the low-resolution image to obtain the manipulation indicated by the driving signal. A high-frequency residual image is computed from the original high-resolution image. A high-frequency manipulated residual image is generated using the motion field. A high-resolution manipulated image is outputted by combining the high-frequency manipulated residual image and a low-frequency manipulated image generated from the low-resolution manipulated image by up-sampling.