Abstract: An image processing method configured to correct a captured image using a neural network includes a first step of determining an inversion axis for a partial image or a filter of the neural network according to a position in the captured image of the partial image that is part of the captured image, a second step of determining a positional relationship among pixels of color components in an image input to the neural network corresponding to the inversion axis, and a third step of generating a corrected image obtained by correcting an input image by processing, using the neural network, the input image generated from the captured image based on the positional relationship.

Abstract: An image processing method includes the steps of acquiring a first image obtained by imaging an object space through a first pupil in an optical system, and a second image obtained by imaging the object space through a second pupil different from the first pupil in the optical system, and generating a blur reshaped image in which a blur caused by a defocus is reshaped based on the first image and the second image using a neural network.

Abstract: An image processing apparatus includes an image acquirer configured to acquire a first image that is generated by image capturing and includes information of light intensity distributions obtained with a plurality of different exposures provided with transmittance distributions that are different from one another, an information acquirer configured to acquire information on the transmittance distributions, and an image estimator configured to separate image signals corresponding to two or more exposures among the plurality of exposures from the first image in the same pixel based on an estimation using the information on the transmittance distributions, and to generate two or more second images using the separated image signals.

Abstract: An image processing method of highly accurately correcting an optical degradation of an image pickup apparatus while also suppressing adverse effects is provided. The image processing method includes a step of obtaining a partial area of an input image shot by using an image pickup apparatus, a step of obtaining previously learnt correction information varying in accordance with a position of the partial area, and a correction step of generating a corrected partial area in which an optical degradation of the partial area caused by the image pickup apparatus is corrected by using the partial area and the correction information.

Abstract: An image processing method includes the steps of acquiring a training image and a correct image, inputting the training image into a multilayer neural network to generate an output image, performing a gamma correction for each of the correct image and the output image and calculating an error between the correct image after the gamma correction and the output image after the gamma correction, and updating a network parameter of the neural network using the error.

Abstract: The image processing apparatus 104 includes a processor performing a noise reduction on at least part of an input image produced by image capturing using an image capturing system 101, 102, and an acquirer acquiring first information on an optical characteristic of the image capturing system. The optical characteristic indicates a factor that degrades information of an object space in the image capturing of the input image. A processor changes a process of the noise reduction depending on the first information.

Abstract: An image capturing apparatus acquires a light intensity distribution in an object space through image capturing using a plurality of pixels, sets a coefficient distribution that is a distribution of a coefficient corresponding to each of the plurality of pixels and is to be applied to the light intensity distribution, acquires object space information that is information on the object space and different from the light intensity distribution, sets first pixel group and second pixel groups that are different from each other in the plurality of pixels based on the object space information, and generates a combined image by combining a plurality of light intensity distributions obtained by applying the coefficient distributions to each of the plurality of pixels obtained with a plurality of image capturing conditions.

Abstract: The image capturing apparatus includes an image sensor including multiple pixels each including at least one photoelectric convertor and a transmittance distribution provider to provide, to at least part of the multiple pixels, a transmittance distribution depending on an incident angle of light. The multiple pixels includes an adjacent pixel group that includes a first pixel and second pixels adjacent to the first pixel, the adjacent pixel group including pixels to which mutually different four or more kinds of the transmittance distributions are provided. The apparatus produces, by using an input image produced from an output of the image sensor and information on the transmittance distributions, an output image corresponding to an optical observation image observable when an object space is observed from a partial area of a pupil of an imaging optical system.

Abstract: An image processing apparatus includes an acquirer which acquires a blurred image and a generator which acquires a first blur estimation area of at least a part of the blurred image to generate an estimated blur, and the generator determines a second blur estimation area based on a signal variation in the first blur estimation area, generates the estimated blur by performing iterative calculation processing that repeats correction processing and estimation processing, and determines the second blur estimation area having a signal variation larger than the signal variation in the first blur estimation area when the signal variation in the first blur estimation area is smaller than a first threshold value.

Abstract: An image processing apparatus includes an acquirer which acquires a blurred image, and a generator which acquires a blur estimation area of at least a part of the blurred image to generate an estimated blur based on the blur estimation area, and the generator generates the estimated blur by performing iterative calculation processing that repeats correction processing and estimation processing, the correction processing correcting a blur included in information relating to a signal in the blur estimation area to generate information relating to a correction signal, and the estimation processing estimating a blur based on the information relating to the signal and the information relating to the correction signal, and generates, as the information relating to the correction signal, information relating to a plurality of correction signals by using a plurality of different calculation expressions in at least one correction processing during the iterative calculation processing.

Abstract: An image processing method includes the steps of acquiring a first image shot by using a compound eye image pickup apparatus, acquiring image capturing condition information of the first image, acquiring, depending on the image capturing condition information, optical characteristic information of a plurality of optical systems having a plurality of focal lengths different from each other in the compound eye image pickup apparatus, correcting a deterioration of the first image caused by shooting the first image based on the optical characteristic information, and generating a second image based on information of a position and an angle of a ray obtained from the first image whose deterioration is corrected, and the optical characteristic information contains at least one of information related to aberrations and information related to peripheral illumination of the optical systems.

Abstract: An image processing method of highly accurately correcting an optical degradation of an image pickup apparatus while also suppressing adverse effects is provided. The image processing method includes a step of obtaining a partial area of an input image shot by using an image pickup apparatus, a step of obtaining previously learnt correction information varying in accordance with a position of the partial area, and a correction step of generating a corrected partial area in which an optical degradation of the partial area caused by the image pickup apparatus is corrected by using the partial area and the correction information.

Abstract: An image processing apparatus includes an image acquirer configured to acquire a first image that is generated by image capturing and includes information of light intensity distributions obtained with a plurality of different exposures provided with transmittance distributions that are different from one another, an information acquirer configured to acquire information on the transmittance distributions, and an image estimator configured to separate image signals corresponding to two or more exposures among the plurality of exposures from the first image in the same pixel based on an estimation using the information on the transmittance distributions, and to generate two or more second images using the separated image signals.

Abstract: An image processing apparatus capable of generating a plurality of output images having different focus positions by reconstructing an input image, includes a storage unit storing image pickup condition information, and an image processing unit generating the output image from the input image using the image pickup condition information, and the image processing unit obtains the input image that is information of an object space viewed from a plurality of viewpoints that is obtained via an imaging optical system and an image pickup element having a plurality of pixels, calculates an average pixel value of a pixel group of the input image of the same region, and substitutes each pixel value of the pixel group by the average pixel value, and performs combination such that the pixels substituted by the average pixel value are shifted from each other to generate the output image.

Abstract: An image capturing apparatus acquires a light intensity distribution in an object space through image capturing using a plurality of pixels, sets a coefficient distribution that is a distribution of a coefficient corresponding to each of the plurality of pixels and is to be applied to the light intensity distribution, acquires object space information that is information on the object space and different from the light intensity distribution, sets first pixel group and second pixel groups that are different from each other in the plurality of pixels based on the object space information, and generates a combined image by combining a plurality of light intensity distributions obtained by applying the coefficient distributions to each of the plurality of pixels obtained with a plurality of image capturing conditions.

Abstract: The image processing apparatus 104 includes a processor performing a noise reduction on at least part of an input image produced by image capturing using an image capturing system 101, 102, and an acquirer acquiring first information on an optical characteristic of the image capturing system. The optical characteristic indicates a factor that degrades information of an object space in the image capturing of the input image. A processor changes a process of the noise reduction depending on the first information.

Abstract: An image pickup apparatus includes an imaging optical system, an image pickup element including a plurality of pixels, a lens array configured such that rays from the same position on an object plane are incident on pixels of the image pickup element different from each other depending on a pupil region of the imaging optical system through which the ray passes, an image processing unit configured to perform image processing for the input image acquired by the image pickup element to generate the output images, and a control unit configured to drive the imaging optical system to perform focus control, the image processing unit is configured to acquire information on a refocus control range, and the control unit is configured to perform the focus control based on the information on the refocus control range.

Abstract: The present invention is directed to an image processing device including: a unit configured to extract, as target data, a set of signal values of pixels included in a target region including a target pixel in an input image; a unit configured to obtain reference data for each of a plurality of reference pixels corresponding to the target pixel; a unit configured to generate transformation reference data by transforming the reference data; a unit configured to determine a weight for each of a plurality of transformation reference pixels based on the target data and the reference data; and a unit configured to generate an output pixel corresponding to the target pixel by using a signal value calculated based on the transformation reference pixels and the weights.

Abstract: An image pickup apparatus (10) capable of generating a plurality of output images having different focus positions by reconstructing an input image, the image pickup apparatus includes an input image obtaining unit (100) configured to obtain the input image, an image processing unit (105) configured to generate a display image from the input image, and a display unit (106) configured to display the display image, and the image processing unit (105) obtains a focus control range in which a focus position is controllable, and generates the display image including information on the focus control range by using at least a part of the input image.

Abstract: An image pickup apparatus includes an image sensor, and a controller configured to drive a focus unit so that a first position is focused, to acquire first information on a first refocusable range that is available in an image that is captured while the first position is being focused by the focus unit, to acquire second information on a second refocusable range that contains the first position, to acquire information on a second position on which the focus unit is to be focused so as to provide the second refocusable range using the first information and the second information, and make the image sensor capture an image while the second position is being focused by the focus unit.