A device includes a memory and one or more processors. The memory is configured to store instructions. The one or more processors are configured to execute the instructions to apply a neural network to a first image to generate an enhanced image. The one or more processors are also configured to execute the instructions to adjust at least a portion of a high-frequency component of the enhanced image based on a motion compensation operation to generate an adjusted high-frequency image component. The one or more processors are further configured to execute the instructions to combine a low-frequency component of the enhanced image and the adjusted high-frequency image component to generate an adjusted enhanced image.

An endoscopic camera device having an optical assembly; a first image sensor in optical communication with the optical assembly, the first image sensor receiving a first exposure and transmitting a first low dynamic range image; a second image sensor in optical communication with the optical assembly, the second image sensor receiving a second exposure and transmitting a second low dynamic range image, the second exposure being higher than the first exposure; and a processor for receiving the first low dynamic range image and the second low dynamic range image; wherein the processor is configured to combine the first low dynamic range image and the second dynamic range image into a high dynamic range image using a luminosity value derived as a preselected percentage of a cumulative luminosity distribution of at least one of the first low dynamic range image and the second low dynamic range image.

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.

A method includes obtaining multiple images of a scene using at least one red-green-blue-white (RGBW) image sensor. The method also includes generating multi-channel frames at different exposure levels from the images. The method further includes estimating motion across exposure differences between the different exposure levels using a white channel of the multi-channel frames as a guidance signal to generate multiple motion maps. The method also includes estimating saturation across the exposure differences between the different exposure levels to generate multiple saturation maps. The method further includes using the generated motion maps and saturation maps to recover saturations from the different exposure levels and generate a saturation-free RGBW frame. In addition, the method includes processing the saturation-free RGBW frame to generate a final image of the scene.

A first aspect of the invention provides a method of training a neural network for capturing volumetric video, comprising: generating a 3D model of a scene; using the 3D model to generate a high fidelity depth map; capturing a perceived depth map of the scene, having a field of view that is aligned with a field of view of the high fidelity depth map; and training the neural network based on the high fidelity depth map and the perceived depth map, wherein the high fidelity depth map has a higher fidelity to the scene than the perceived depth map has.

In an embodiment, a method (100) is described. The method comprises receiving (102) data corresponding to a plurality of radiographic imaging slices of a body. The method further comprises determining (104) a position of a needle inserted in the body. The determination is based on combining information from at least one of the radiographic imaging slices comprising an indication of a first portion of the needle outside the body and at least one other of the radiographic imaging slices comprising an indication of a second portion of the needle inside the body. A combined needle region is generated by merging data corresponding to a position of the first portion of the needle outside the body with data corresponding to a position of the second portion of the needle inside the body. The method further comprises generating (106) display data for providing a visual representation of the needle in an image of the body in combination with a visual representation of at least the first and second portions of the needle superimposed on the image. The image is in a plane that is digitally tilted with respect to a plane parallel to the plurality of radiographic imaging slices.

Disclosed are a display apparatus, a method for synthesizing images of a moving object and a device. The display apparatus includes: a display panel, including a main display region and a transparent display region at least partially provided in the main display region; and an image acquisition component, provided on a non-display side of the transparent display region, and configured to acquire a light incident from a display side of the transparent display region and penetrating the transparent display region. The image acquisition component is configured to continuously acquire, at a predetermined time interval, light of a moving object transmitting through the transparent display region, to obtain n images, diffraction spots of at least two of the n images are different. The image acquisition component is further configured to merge the n images to obtain m synthetic image(s) to eliminate or weaken diffraction spots in the n images.

An electronic device includes a UDC and a UDC controller configured to determine an optimal number of frames required to be captured for a scene to compensate at least one parameter to optimize an output media using the UDC, obtain a multi-frame fusion media by performing at least one multi-frame fusion on the determined optimal number of frames, perform a light source spread correction on the multi-frame fusion media, and optimize the output media based on the light source spread correction on the multi-frame fusion media.

Provided are a video conversion method, an electronic device and a non-transitory computer readable storage medium. The implementation scheme is as follows: a to-be-converted SDR video is acquired; one frame is extracted from the to-be-converted SDR video to serve as a current SDR image, the current SDR image is input into a parameter predictor and a generator, and an adjustment parameter corresponding to the current SDR image is output from the parameter predictor; the adjustment parameter corresponding to the current SDR image is input into the generator, and an HDR image corresponding to the current SDR image is output from the generator; and the operation described above is repeatedly performed until frames are converted into HDR images each of which corresponds to a respective frame of the frames; and a corresponding HDR video is generated based on the HDR images corresponding to the frames.

A method of controlling a mobile robot includes a first basic learning process of generating a first basic map based on environment information acquired in a traveling process, a second basic learning process of generating a second basic map based on environment information acquired in a separate traveling process, and a merging process of merging the first basic map and the second basic map to generate a merged map.