Sony Corporation

Sony Semiconductor Solutions Corporation

Nippon Metal Industry Co., Ltd.

Sony Group Corporation

Hiroshi Arai

Norimitsu Okiyama

Yuji Matsui

Kenichiro Nishi

Takeshi Masuura

Satoshi Takashima

Yuriko Ohtsuka

Hiroaki Takahashi

Masahiro Sasaki

Seiichi Takeda

Kazuhisa Yokoyama

Jyou Tanioka

Yoshio Sawamura

Hidetoshi Abo

Hiroshi Yoshimura

Naoto Toyama

Yumi Nagaoka

Kazuo Mashimo

Tadahiko Murakami

A signal processing device according to the present technology includes a feature quantity extraction unit including a neural network and trained to extract a feature quantity for a specific event with respect to an input signal from a sensor, and a correction unit that performs correction of the input signal on the basis of the feature quantity extracted by the feature quantity extraction unit.

Signal processing device and signal processing method for correcting input signal from sensor

A signal processing device includes an approximate curved surface conversion unit that includes a first stacked autoencoder pretrained on the basis of learning input data constituted by coordinate data acquired for each of multiple elements of an object, and that obtains approximate curved surface data indicating an approximate curved surface of the object in an intermediate layer of the first stacked autoencoder, on the basis of input data constituted by the coordinate data acquired for each of the elements, and a geometric modulation processing unit that includes a second stacked autoencoder having learned by machine learning on the basis of learning input data constituted by the approximate curved surface data and on the basis of training data constituted by a result obtained by coordinate conversion for each of the elements in a geometric modulation process performed for the object, and performs the geometric modulation process.

Signal processing device and signal processing method

A signal processing device according to the present technology includes a stacked auto encoder that processes an input signal from a sensor, a control line associated learner including a neural network and subjected to control line associated learning for performing learning by associating different event aspects related to a specific event with values of different control lines with a feature quantity obtained in an intermediate layer of the stacked auto encoder after pretraining as an input, and a refactorer that obtains a difference between a first output that is an output of the control line associated learner when a first value is given to the control line, and a second output that is an output of the control line associated learner when a second value different from the first value is given to the control line.

Signal processing device, signal processing method, and parameter search method

An information processing apparatus includes a control unit that creates a plurality of learning information items including an input image and a teacher image as an expected value by image-processing the input image in accordance with a scenario described with a program code, and supplies the created plurality of learning information items to a machine learning module that composes an image processing algorithm by machine learning.

Information processing apparatus and information processing method to generate image processing algorithm based on machine learning process

Posture information corresponding to a line of an input image is calculated. Then, a first virtual line, which is a line in a line direction of the input image for which the posture information calculated is common, is rotated according to the posture information to obtain a second virtual line on the output image. For each pixel of the output image, a corresponding second virtual line is calculated, and posture information is obtained on the basis of the corresponding second virtual line. Then, for coordinates of each pixel of the output image, the reference coordinates on the input image corresponding to the pixel coordinates is calculated using the associated posture information. By cutting out the output image from the input image on the basis of the reference pixel, highly accurate distortion correction can be realized even in a case where the camera posture differs depending on the line.

Signal processing device, imaging device, and signal processing method

This information processing apparatus includes: a mechanical learning module that is a multiplier-accumulator circuit including a plurality of deep-learning neural network blocks hierarchized and synthesizes an algorithm having a filter characteristic for image processing by mechanical learning, the image processing being a target; a debug module that analyzes a digital filter characteristic in each of the deep-learning neural network blocks on the basis of input and output data in the mechanical learning module; and a grading module that develops an output image from an input image by performing gain control on the basis of a result of analysis of the debug module.

Information processing apparatus and information processing method

An imaging apparatus includes an imaging unit and a transmission unit. The imaging unit is configured to capture two images that are different from each other by a predetermined amount of an optical distance (focus) between an objective lens and an imaging device having a first resolution. The transmission unit is configured to transmit the captured images.

Imaging apparatus and imaging method

Information processing apparatus and information processing method to generate image processing algorithm based on achine learning process

Imaging apparatus and imaging method with error corrected interpolation frame generation

Imaging apparatus and imaging method for generating increased resolution images, hyperspectral images, steroscopic images, and/or refocused images

There is provided an image processing method includes: separating an image taken at a predetermined frame rate into a first frame and at least one second frame other than the first frame at intervals of 1/n, where n is an integer of 2 or larger; calculating a low-frequency-component difference between the separated at least one second frame and first frame; performing signal processing designated by a user on the first frame; decompressing, using a low-frequency component in the first frame being subjected to the signal processing and the low-frequency-component difference, a low-frequency component in the at least one second frame being approximately subjected to the signal processing; and decompressing, using the decompressed low-frequency component in the at least one second frame and a high-frequency component therein, the at least one second frame being approximately subjected to the signal processing.

Image processing method, encoding device, decoding device, and image processing apparatus

An encoder includes: an image decomposition unit configured to input an image and decompose the input image into a first component capable of being used as a monitoring image and a second component other than the first component by frequency decomposition, the input image being restored by coupling the first component and the second component to each other; a first transmission unit configured to output the decomposed first component as a first-channel transmission image; a compression coding unit configured to compression-code the second component; and a second transmission unit configured to output the compression-coded second component as a second-channel transmission image.

Encoder, decoder, encoder system, decoder system, transmission adapter, encoding method, decoding method, and imaging apparatus

An image processing apparatus includes a characteristic point extracting section and a signal processing section. The characteristic point extracting section is configured to analyze a frequency component of an inputted image and extracts a plurality of pixels characterizing the image as a plurality of characteristic points. The signal processing section is configured to perform signal processing for the plurality of extracted pixels corresponding to the respective characteristic points, produces an interpolated value based on a result of the signal processing, and rewrites a value of at least one pixel other than the characteristic points to the interpolated value.

Image processing apparatus and method of processing image

A video-processing apparatus contains a region-setting-processing unit that sets a plurality of zoom regions each enlarging a part of an input video image and displaying the enlarged part of the input video image and creates viewpoint information indicating the set zoom regions. The input video image has a higher resolution than a display resolution. The video-processing apparatus also contains a video-data-extracting unit that extracts video data from video data of the input video image for every set zoom region, a video-data-converting unit that downconverts the extracted video data to video data of the display resolution, and an output unit that outputs the downconverted video data and the viewpoint information with the viewpoint information corresponding to the downconverted video data.

Growing community of inventors

Kanagawa, Japan

Hiroshi Arai