A method for processing images acquired by a multi-spectral RGB-NIR (red/green/blue/near infra-red) sensor includes receiving a RGB-NIR digital image from a multi-spectral RGB-NIR sensor, interpolating an NIR contribution to each R, G and B pixel value, wherein an NIR image is obtained, subtracting the NIR contribution from each R, G and B pixel value in the RGB-NIR digital image wherein a decontaminated RGB-NIR image is obtained, constructing a red, green and blue (RGB) Bayer image from the decontaminated RGB-NIR image, and processing the Bayer image wherein a full color image is obtained. The RGB-NIR digital image includes red (R) pixels, green (G) pixels, blue (B) pixels, and NIR pixels, and every other row in the RGB-NIR digital image includes NIR pixels that alternate with green pixels, and every other row in the RGB-NIR digital image includes green pixels that alternate with red and blue pixels.

An information processing apparatus includes: a memory; and a processor configured to: execute genetic processing on class classification programs of a set of class classification programs; acquire a distance between each learning data and an identification boundary regarding evaluation identifiers created by using first class learning data belonging to a first class and second class learning data belonging to a second class according to the class classification program; calculate a statistic amount of a distribution of a distance to the identification boundary for each of the first class learning data and the second class learning data; define a fitness calculation equation based on the statistic amounts of the first class learning data and the second class learning data; calculate the fitness of the class classification program; and determine whether or not to replace the class classification program with one of class classification programs of the set depending on the fitness.

A video frame interpolation method and device, and a computer-readable storage medium are described. The method includes: inputting at least two image frames into a video frame interpolation model to obtain at least one frame-interpolation image frame, training the initial model using a first loss to obtain a reference model, copying the reference model to obtain three reference models with shared parameters, selecting different target sample images according to a preset rules to train the first/second reference model to obtain a first/second frame-interpolation result; selecting third target sample images from the first/second frame-interpolation result to train the third reference model to obtain the frame-interpolation result, obtaining a total loss of the first training model based on the frame-interpolation result and the sample images, adjusting parameters of the first training model based on the total loss, and using a parameter model via a predetermined number of iterations as the video frame interpolation model.

An information processing apparatus includes: a memory; and a processor configured to: execute genetic processing on class classification programs of a set of class classification programs; acquire a distance between each learning data and an identification boundary regarding evaluation identifiers created by using first class learning data belonging to a first class and second class learning data belonging to a second class according to the class classification program; calculate a statistic amount of a distribution of a distance to the identification boundary for each of the first class learning data and the second class learning data; define a fitness calculation equation based on the statistic amounts of the first class learning data and the second class learning data; calculate the fitness of the class classification program; and determine whether or not to replace the class classification program with one of class classification programs of the set depending on the fitness.

An image conversion method includes: dividing a low-resolution image to generate a plurality of low-resolution image patches having N (N being a natural number of 2 or greater) number of pixels; classifying the low-resolution image patches into a plurality of image categories; converting the low-resolution image patches by using a conversion kernel for each image patch corresponding to its image category among prestored conversion matrices, to generate high-resolution image patches having L (L being a natural number) number of pixels, which is less than the N number of pixels; and arranging the high-resolution image patches to generate a high-resolution image. According to the image conversion method, an operation amount can be reduced compared with a conventional technique, and thus the image conversion method can be implemented in hardware having low complexity. A high quality and high-resolution image can be generated regardless of characteristic information of an image signal.

Systems and methods for generating interpolated images are disclosed. In examples, image features are extracted from a first image and a second image; such image features may be warped using first and second plurality of parameters. A first candidate intermediate frame may be generated based on the warped first features and the warped second features. Multi-scale features associated with the image features extracted from the first image and the second image may be obtained and warped using the first and second plurality of parameters. A second candidate intermediate frame may be generated based on the warped first multi-scale features and the warped second multi-scale features. By blending the first candidate intermedia frame with the second candidate intermediate frame, an interpolated image may be generated.

An image conversion method includes: dividing a low-resolution image to generate a plurality of low-resolution image patches having N (N being a natural number of 2 or greater) number of pixels; classifying the low-resolution image patches into a plurality of image categories; converting the low-resolution image patches by using a conversion kernel for each image patch corresponding to its image category among prestored conversion matrices, to generate high-resolution image patches having L (L being a natural number) number of pixels, which is less than the N number of pixels; and arranging the high-resolution image patches to generate a high-resolution image. According to the image conversion method, an operation amount can be reduced compared with a conventional technique, and thus the image conversion method can be implemented in hardware having low complexity. A high quality and high-resolution image can be generated regardless of characteristic information of an image signal.

An electronic apparatus includes a memory configured to store an input image and at least one processor configured to obtain two consecutive frames of the input image as input frames; obtain a first interpolation frame of the input frames and a first confidence corresponding to the first interpolation frame based on a first interpolation method; obtain a second interpolation frame of the input frames and a second confidence corresponding to the second interpolation frame based on a second interpolation method that is different from the first interpolation method; obtain weights corresponding to the first interpolation frame and the second interpolation frame, based on the first confidence and the second confidence, respectively; and obtain an output image based on the weights.

To enable a high-quality image process while reducing power consumption, an image processing apparatus is characterized by acquiring a plurality of image frames constituting a moving image; detecting a feature quantity of the image frame; determining the number of image frames to be used for the image process of the image frame, based on the detected feature quantity; and performing the image process to a process-object frame, by using the image frames corresponding to the determined number of image frames.

A method for processing images acquired by a multi-spectral RGB-NIR (red/green/blue/near infra-red) sensor includes receiving a RGB-NIR digital image from a multi-spectral RGB-NIR sensor, interpolating an NIR contribution to each R, G and B pixel value, wherein an NIR image is obtained, subtracting the NIR contribution from each R, G and B pixel value in the RGB-NIR digital image wherein a decontaminated RGB-NIR image is obtained, constructing a red, green and blue (RGB) Bayer image from the decontaminated RGB-NIR image, and processing the Bayer image wherein a full color image is obtained. The RGB-NIR digital image includes red (R) pixels, green (G) pixels, blue (B) pixels, and NIR pixels, and every other row in the RGB-NIR digital image includes NIR pixels that alternate with green pixels, and every other row in the RGB-NIR digital image includes green pixels that alternate with red and blue pixels.