Abstract: Fabrication processing is executed in a chip of an image sensor. An image capturing device includes an image capturing unit (11) mounted on a vehicle and configured to generate image data by performing image capturing of a peripheral region of the vehicle, a scene recognition unit (214) configured to recognize a scene of the peripheral region based on the image data, and a drive control unit (12) configured to control drive of the image capturing unit based on the scene recognized by the scene recognition unit.

Abstract: A stacked light receiving sensor according to one embodiment includes a first substrate in a first layer, a second substrate joined with the first substrate and formed in a second layer, and a third substrate joined with the second substrate and formed in a third layer. An analog circuit reads a pixel signal from a pixel array. A logic circuit is connected to the analog circuit and outputs the pixel signal. A processing section executes processing based on a neural network computing model, on data based on the pixel signal. The pixel array is disposed on the first layer. The analog circuit is disposed on any one or more of the first to third layers. The logic circuit, the processing section, and a memory are disposed on any one or more of the second and third layers.

Abstract: The present invention suppresses an increase in processing time and power consumption that is associated with function implementation. An electronic device according to an embodiment comprises: an imaging unit (11) that generates image data; a processing unit (14) that executes neural network calculation model-based processing with respect to image data-based data that is read out from the imaging unit; a function execution unit (12) that performs a predetermined function on the basis of the results of the processing; and a detection unit (32) that detects displacement, wherein the processing unit executes the processing when the detection unit detects displacement.

Abstract: An imaging apparatus includes: an imaging unit that has a pixel region in which a plurality of pixels is arranged and that reads and outputs a pixel signal from the pixels included in the pixel region; a unit-of-readout controller that controls a unit of readout set as a part of the pixel region: a recognition unit that has learned training data for each of units of readout; a first output unit that outputs a recognition result of recognition performed by the recognition unit for each of the units of readout; a second output unit that outputs the pixel signal to a subsequent stage; and a trigger generator that generates a trigger signal for controlling a first timing at which the first output unit outputs the recognition result and a second timing at which the second output unit outputs the pixel signal.

Abstract: Advanced processing is performed in a chip. A stacked light-receiving sensor according to an embodiment includes a first substrate (100, 200, 300), a second substrate (120, 320) bonded to the first substrate, and connection wiring (402) bonded to the second substrate. The first substrate includes a pixel array (101) in which a plurality of unit pixels are arranged in a two-dimensional matrix. The second substrate includes a converter (17A) configured to convert an analog pixel signal output from the pixel array to digital image data and a processing unit (14) configured to perform a process for data based on the image data. At least a part of the converter is disposed on a first side in the second substrate. The processing unit is disposed on a second side opposite to the first side in the second substrate. The connection wiring is attached to a side other than the second side in the second substrate.

Abstract: An imaging apparatus includes: an imaging unit that has a pixel region in which a plurality of pixels is arranged and reads and outputs a pixel signal from the pixels included in the pixel region; a unit-of-readout controller that controls a unit of readout set as a part of the pixel region; a first unit-of-readout setting unit that sets a first unit of readout that is used for reading out the pixel signal from the pixel region for a recognition process that has learned training data for each of the units of readout; a second unit-of-readout setting unit that sets a second unit of readout that is used for reading out the pixel signal from the pixel region so as to output the pixel signal to a subsequent stage; and a mediation unit that performs mediation between the first unit of readout and the second unit of readout.

Abstract: A solid-state imaging device includes: an imager (11) configured to acquire image data; a processing unit (14) configured to execute, on the image data or data based on the image data, processing of extracting a specific region based on a neural network calculation model; and an output unit (16) configured to output image data fabricated based on the specific region or the image data read from the imager based on the specific region.

Abstract: An imaging apparatus according to an embodiment includes: an imaging unit having a pixel region in which a plurality of pixels is arranged; a readout controller that controls readout of pixel signals from pixels included in the pixel region; a first unit-of-readout setting unit that sets a unit of readout as a part of the pixel region, for which the readout controller performs readout of the pixel signal; an image output unit that outputs a first image based on the pixel signal read out from the unit of readout to a subsequent stage; a second unit-of-readout controller that controls the unit of readout in which the readout controller performs readout of the pixel signal; and a recognition unit that learns training data for each of the units of readout, performs a recognition process on the pixel signal for each of the units of readout, and outputs a recognition result.

Abstract: A stacked light receiving sensor according to one embodiment includes a first substrate in a first layer, a second substrate joined with the first substrate and formed in a second layer, and a third substrate joined with the second substrate and formed in a third layer. An analog circuit reads a pixel signal from a pixel array. A logic circuit is connected to the analog circuit and outputs the pixel signal. A processing section executes processing based on a neural network computing model, on data based on the pixel signal. The pixel array is disposed on the first layer. The analog circuit is disposed on any one or more of the first to third layers. The logic circuit, the processing section, and a memory are disposed on any one or more of the second and third layers.

Abstract: Provided are an information processing device, an information processing method, and an information processing program capable of reducing a processing load of convolution processing in a convolutional neural network (CNN). An information processing device (1) according to the present disclosure includes a setting unit (51) and a control unit (52). The setting unit (51) sets exposure time of each of imaging pixels in an imaging unit (2), which includes a plurality of imaging pixels arrayed two-dimensionally, to exposure time corresponding to a convolution coefficient of a first layer of a CNN. The control unit (52) causes transfer of signal charges from imaging pixels, which have been exposed, to a floating diffusion (FD), thereby performing convolution processing.

Abstract: An imaging apparatus includes: an imaging unit that has a pixel region in which a plurality of pixels is arranged and reads and outputs a pixel signal from the pixels included in the pixel region; a unit-of-readout controller that controls a unit of readout set as a part of the pixel region; a first unit-of-readout setting unit that sets a first unit of readout that is used for reading out the pixel signal from the pixel region for a recognition process that has learned training data for each of the units of readout; a second unit-of-readout setting unit that sets a second unit of readout that is used for reading out the pixel signal from the pixel region so as to output the pixel signal to a subsequent stage; and a mediation unit that performs mediation between the first unit of readout and the second unit of readout.

Abstract: An imaging apparatus includes: an imaging unit that has a pixel region in which a plurality of pixels is arranged and that reads and outputs a pixel signal from the pixels included in the pixel region; a unit-of-readout controller that controls a unit of readout set as a part of the pixel region: a recognition unit that has learned training data for each of units of readout; a first output unit that outputs a recognition result of recognition performed by the recognition unit for each of the units of readout; a second output unit that outputs the pixel signal to a subsequent stage; and a trigger generator that generates a trigger signal for controlling a first timing at which the first output unit outputs the recognition result and a second timing at which the second output unit outputs the pixel signal.

Abstract: Advanced processing is performed in a chip. A stacked light-receiving sensor according to an embodiment includes a first substrate (100, 200, 300) and a second substrate (120, 320) bonded to the first substrate. The first substrate includes a pixel array (101) in which a plurality of unit pixels are arranged in a two-dimensional matrix. The second substrate includes a converter (17) configured to convert an analog pixel signal output from the pixel array to digital image data and a processing unit (15) configured to perform a process based on a neural network calculation model for data based on the image data. At least a part of the converter is arranged on a first side in the second substrate. The processing unit is arranged on a second side opposite to the first side in the second substrate.

Abstract: A solid-state imaging device (10) includes an imager (11) configured to acquire image data, a processing unit (14) configured to perform a process based on a neural network calculation model for data based on the image data acquired from the imager, and a control unit (12) configured to switch between a first process mode of performing a first process at a first frame rate and, based on a result of the first process, a second process mode of performing a second process at a second frame rate.

Abstract: Fabrication processing is executed in a chip of an image sensor. An image capturing device includes an image capturing unit (11) mounted on a vehicle and configured to generate image data by performing image capturing of a peripheral region of the vehicle, a scene recognition unit (214) configured to recognize a scene of the peripheral region based on the image data, and a drive control unit (12) configured to control drive of the image capturing unit based on the scene recognized by the scene recognition unit.

Abstract: An imaging apparatus according to an embodiment includes: an imaging unit having a pixel region in which a plurality of pixels is arranged; a readout controller that controls readout of pixel signals from pixels included in the pixel region; a first unit-of-readout setting unit that sets a unit of readout as a part of the pixel region, for which the readout controller performs readout of the pixel signal; an image output unit that outputs a first image based on the pixel signal read out from the unit of readout to a subsequent stage; a second unit-of-readout controller that controls the unit of readout in which the readout controller performs readout of the pixel signal; and a recognition unit that learns training data for each of the units of readout, performs a recognition process on the pixel signal for each of the units of readout, and outputs a recognition result.

Abstract: A solid-state imaging device includes an imager configured to acquire image data, a processing unit configured to perform a process based on a neural network calculation model for data based on the image data acquired from the imager, and a control unit configured to switch between a first process mode of performing a first process at a first frame rate and, based on a result of the first process, a second process mode of performing a second process at a second frame rate.

Abstract: Fabrication processing is executed in a chip of an image sensor. An image capturing device includes an image capturing unit (11) mounted on a vehicle and configured to generate image data by performing image capturing of a peripheral region of the vehicle, a scene recognition unit (214) configured to recognize a scene of the peripheral region based on the image data, and a drive control unit (12) configured to control drive of the image capturing unit based on the scene recognized by the scene recognition unit.

Abstract: An imaging apparatus according to an embodiment includes: an imaging unit (10) having a pixel region in which a plurality of pixels is arranged; a readout controller (11) that controls readout of pixel signals from pixels included in the pixel region; a unit-of-readout controller (123) that controls a unit of readout that is set as a part of the pixel region and for which the readout controller performs the readout; and a recognition unit (14) that has learned training data for each of the units of readout. The recognition unit performs a recognition process on the pixel signal for each of the units of readout, and outputs a recognition result which is a result of the recognition process.

Abstract: Provided are an information processing device, an information processing method, and an information processing program capable of reducing a processing load of convolution processing in a convolutional neural network (CNN). An information processing device (1) according to the present disclosure includes a setting unit (51) and a control unit (52). The setting unit (51) sets exposure time of each of imaging pixels in an imaging unit (2), which includes a plurality of imaging pixels arrayed two-dimensionally, to exposure time corresponding to a convolution coefficient of a first layer of a CNN. The control unit (52) causes transfer of signal charges from imaging pixels, which have been exposed, to a floating diffusion (FD), thereby performing convolution processing.