Abstract: A joint position estimation device including a memory, and a processor connected to the memory. The processor executes a process including estimating, by a first DNN for which a first parameter determined by learning of the first DNN has been set, a body part region of the animal with respect to input image to be processed; and estimating, by the second DNN for which a second parameter determined by learning of the second DNN has been set, a first joint position and a second joint position in each of the body part region estimated by the first DNN and a plural body parts region in which a plurality of the body part regions are connected.

Abstract: An information processing device includes, a memory, and a processor coupled to the memory and the processor configured to, acquire a neural network, divide the neural network into divided neural networks, identify, based on input-output characteristics of the divided neural networks, parameters of each of polynomial neural networks corresponding to each of the divided neural networks, and output another neural network generated by linking the identified polynomial neural networks.

Abstract: An arithmetic processing apparatus, includes a memory; and a processor coupled to the memory and the processor configured to: estimate an optical flow between consecutive frames of a moving image, and generate a sound wave corresponding to a target moving image, based on an optical flow estimated for the target moving image, and a deep neural network that generates a sound wave corresponding to an input optical flow and that has performed learning, by associating with time, an optical flow estimated for a moving image with sound wave and a sound wave of the moving image with sound wave.

Abstract: A non-transitory computer-readable recording medium having stored therein an information processing program for causing a computer to execute a process includes: calculating, by using an image including an object having a plurality of joints and having a plurality of frames consecutive in time, a probability distribution of a joint position of the object in each frame; calculating a transition score by evaluating a consistency of a transition of a joint of the plurality of joints using transition information of the joint among different frames; and estimating a pose of the object from a score of the probability distribution and the transition score.

Abstract: An information processing device includes: a memory; and a processor coupled to the memory and configured to: estimate a first position of a first joint in a two-dimensional plane based on a color image, and estimate a first positional relationship between the first joint and a second joint coupled to the first joint via a portion in the two-dimensional plane; estimate a second position of the first joint in the two-dimensional plane based on a depth image indicating respective depths of the first joint and the second joint, and estimate a second positional relationship between the first joint and the second joint in a three-dimensional space; and estimate a third position of the first joint in the two-dimensional plane and posture of the portion in the three-dimensional space based on the first position, the first positional relationship the second position, and the second positional relationship.

Abstract: A data server includes a controller, a memory that stores first information collected from devices, and master-slave management circuitry that sets whether the data server operates according to one of a master function for managing a data server and a slave function for being managed by the data server. In a case where the master function is set by the master-slave management circuitry, the controller sends to an external client, in response to receiving a data collection request from the external client, information corresponding to the data collection request that is one of data stored in the memory and data received from the other data server. In a case where the slave function is set by the master-slave management circuitry, the controller sends the first information to the other data server in which the master function is set.

Abstract: A part recognition method includes: cutting, by a computer, out a plurality of partial images having different sizes using each of positions of an input image as a reference; calculating a probability that each of the partial images is an image indicating a part; calculating, for each of the positions, a score by integrating the probability for each of the partial images; and recognizing, based on the score for each of the positions, the part from the input image.

Abstract: A joint position estimation device including a memory, and a processor connected to the memory. The processor executes a process including estimating, by a first DNN for which a first parameter determined by learning of the first DNN has been set, a body part region of the animal with respect to input image to be processed; and estimating, by the second DNN for which a second parameter determined by learning of the second DNN has been set, a first joint position and a second joint position in each of the body part region estimated by the first DNN and a plural body parts region in which a plurality of the body part regions are connected.

Abstract: A data server includes a data storage section that stores first information collected from devices; a master-slave management section that sets whether the data server has a master function for managing a data server or a slave function for being managed by the data server; and a configuration management section. In a case where the master function is set by the master-slave management section, the configuration management section sends, in response to receiving a data collection request from an external client, information corresponding to the data collection request from the data storage section or the data server to the client. In a case where the slave function is set by the master-slave management section, the configuration management section sends the first information to the data server in which the master function is set.

Abstract: An information processing device includes: a memory; and a processor coupled to the memory and configured to: estimate a first position of a first joint in a two-dimensional plane based on a color image, and estimate a first positional relationship between the first joint and a second joint coupled to the first joint via a portion in the two-dimensional plane; estimate a second position of the first joint in the two-dimensional plane based on a depth image indicating respective depths of the first joint and the second joint, and estimate a second positional relationship between the first joint and the second joint in a three-dimensional space; and estimate a third position of the first joint in the two-dimensional plane and posture of the portion in the three-dimensional space based on the first position, the first positional relationship the second position, and the second positional relationship.

Abstract: A non-transitory computer-readable recording medium having stored therein an information processing program for causing a computer to execute a process includes: calculating, by using an image including an object having a plurality of joints and having a plurality of frames consecutive in time, a probability distribution of a joint position of the object in each frame; calculating a transition score by evaluating a consistency of a transition of a joint of the plurality of joints using transition information of the joint among different frames; and estimating a pose of the object from a score of the probability distribution and the transition score.

Abstract: A part recognition method includes: cutting, by a computer, out a plurality of partial images having different sizes using each of positions of an input image as a reference; calculating a probability that each of the partial images is an image indicating a part; calculating, for each of the positions, a score by integrating the probability for each of the partial images; and recognizing, based on the score for each of the positions, the part from the input image.

Abstract: An apparatus for controlling a frequency analysis processing includes: a memory; and a processor coupled to the memory and configured to execute a fast Fourier transform process that includes performing a fast Fourier transform operation on data of two groups into which sensor data sensed at a first sampling frequency by a sensor is divided, and execute a change process that includes changing, in a case where results of butterfly operations of the fast Fourier transform operation are similar between the two groups, a sampling frequency at which the sensor operates to a second sampling frequency lower than the first sampling frequency.

Abstract: A symbol detector includes a binarization processor configured to read a symbol image in a uniaxial direction and output a first pixel value or a second pixel value obtained by binarization of a pixel value of each pixel of the symbol image, a calculator configured to calculate a second ratio of a first consecutive value and a second consecutive value obtained before and after the first consecutive value in a case where the first consecutive value and the second consecutive value are switched alternately, and a transmitter configured to output the symbol image in a case where a difference between the calculated second ratio and the first ratio is equal to or lower than a predetermined value, and stop outputting the symbol image in a case where the difference is higher than the predetermined value.

Abstract: An information processing device includes, a memory, and a processor coupled to the memory and the processor configured to, acquire a neural network, divide the neural network into divided neural networks, identify, based on input-output characteristics of the divided neural networks, parameters of each of polynomial neural networks corresponding to each of the divided neural networks, and output another neural network generated by linking the identified polynomial neural networks.

Abstract: In a heart rate estimating apparatus, a sensor generates pulse wave data indicating a pulse wave of a body. A storage unit stores a plurality of state transition models each representing states including a normal state and an abnormal state corresponding to noise, transition probabilities between the states, and output probabilities of a plurality of different symbols, wherein at least one of the transition probabilities and the output probabilities differs between the respective state transition models. An estimating unit generates a symbol string that indicates a time series of changes in the pulse wave from the pulse wave data, calculates a fit between the symbol string and each of the plurality of state transition models, and calculates a heart rate estimated from the pulse wave data based on a state transition model selected in accordance with the fit.

Abstract: An apparatus for controlling a frequency analysis processing includes: a memory; and a processor coupled to the memory and configured to execute a fast Fourier transform process that includes performing a fast Fourier transform operation on data of two groups into which sensor data sensed at a first sampling frequency by a sensor is divided, and execute a change process that includes changing, in a case where results of butterfly operations of the fast Fourier transform operation are similar between the two groups, a sampling frequency at which the sensor operates to a second sampling frequency lower than the first sampling frequency.

Abstract: A symbol detector includes a binarization processor configured to read a symbol image in a uniaxial direction and output a first pixel value or a second pixel value obtained by binarization of a pixel value of each pixel of the symbol image, a calculator configured to calculate a second ratio of a first consecutive value and a second consecutive value obtained before and after the first consecutive value in a case where the first consecutive value and the second consecutive value are switched alternately, and a transmitter configured to output the symbol image in a case where a difference between the calculated second ratio and the first ratio is equal to or lower than a predetermined value, and stop outputting the symbol image in a case where the difference is higher than the predetermined value.

Abstract: A semiconductor device includes: a power supply; a circuit block that has at least one storage element and operates by receiving a power supply voltage from the power supply; a power management unit that controls the power supply to change the power supply voltage; and a storage element monitor circuit that generates a first malfunction signal at a first margin voltage that is higher than a voltage at which the storage element does not normally operate in a case where the power supply voltage lowers, wherein the power management unit controls the power supply so that the power supply voltage does not become lower than the first margin voltage.

Abstract: A circuit including: an input stage that includes a first input unit into which input data is input and a pair of first output units and is driven by a first power-supply voltage; a pair of first gate elements that includes first transistors, and is driven by a clock that includes a second power-supply voltage that is lower than the first power-supply voltage; a first latch circuit that includes a pair of second input units, and is driven by the first power-supply voltage; a pair of second gate elements that includes second transistors, and is driven by an inverted clock of the clock; and a second latch circuit that includes a pair of third input units, and a third output unit that outputs one of a pair of pieces of data, and is driven by the first power-supply voltage.