Methods and apparatus for identifying media content using temporal signal characteristics are disclosed. An example apparatus includes at least one memory, computer readable instructions, and at least one processor to execute the instructions to identify intervals in a media signal; generate interval sums for respective ones of the intervals, a first interval sum of the interval sums based on a sum of magnitudes of first peaks of the media signal that occur between zero crossings of a first interval of the intervals of the media signal; identify second peaks based on the interval sums; and generate a signature representative of the media signal based on the second peaks.

Abnormality detection device includes: processing circuitry performing a process that: extracts a first feature amount using a sliding window of a first time length and a second feature amount using a sliding window of a second time length longer than the first time length; calculates a unit incremental value by dividing a specific value difference subtracting a first specific value in the first feature amount from a second specific value in the second feature amount by a time length difference subtracting the first time length from the second time length; sequentially calculates, for each abnormality detection time length different from each other, a threshold based on the unit incremental value; and sequentially generates, for each abnormality detection time length, a plurality of partial time series having the abnormality detection time lengths from the time series data, and detects an abnormality in the time series data based on those and the threshold.

Systems and methods for detecting patterns in data from a time-series are provided. In one implementation, a method for pattern detection includes obtaining data in a time-series and creating one-dimensional or multi-dimensional windows from the time-series data. The one-dimensional or multi-dimensional windows are created either independently or jointly with the time-series. The method also includes training a deep neural network with the one-dimensional or multi-dimensional windows utilizing historical and/or simulated data to provide a neural network model. Also, the method includes processing ongoing data with the neural network model to detect one or more patterns of a particular category in the ongoing data, and localizing the one or more patterns in time.

A multiple sensor-fusing based interactive training system, including a posture sensor, a sensing module, a computing module, and a display module, is provided. The posture sensor is configured to sense posture data and myoelectric data related to a training action. The sensing module is configured to output limb torque data according to the posture data, and output muscle group activation time data according to the myoelectric data. The computing module is configured to respectively convert the limb torque data and the muscle group activation time data into a moment-skeleton coordinate system and a muscle strength eigenvalue-skeleton coordinate system according to a skeleton coordinate system, perform fusion calculation, calculate evaluation data based on a result of the fusion calculation, and judge that the training action corresponds to a known exercise action according to the evaluation data. The display module is configured to display the evaluation data and the known exercise action.

Introduced here are computer programs and associated computer-implemented techniques for determining reflectance of an image on a per-pixel basis. More specifically, a characterization module can initially acquire a first data set generated by a multi-channel light source and a second data set generated by a multi-channel image sensor. The first data set may specify the illuminance of each channel of the multi-channel light source (which may be able to produce visible light and/or non-visible light), while the second data set may specify the response of each sensor channel of the multi-channel image sensor (which is configured to capture an image in conjunction with the light). Thus, the characterization module may determine reflectance based on illuminance and sensor response. The characterization module may also be configured to determine illuminance based on reflectance and sensor response, or determine sensor response based on illuminance and reflectance.

A natural language processing combination classifier is disclosed, leveraging an entailment classifier and optionally at least one of a pattern matching classifier and a trained machine learning (ML) classifier. Each of the different types of classifiers can be used to identify different categories of matches. For example, a pattern matching (e.g., regular expression) classifier may identify exact matches, an entailment classifier can obtain context-specific classifications based on likelihood of entailment to comparison data, and the SML classifier can obtain potential matches based on large-scale supervised training. The comparison data for the entailment classifier can be generated from small datasets, and can be readily updated without the need for retraining any machine learning models. Different types of classifiers can be processed using different logic to provide a user with the most appropriate response given a circumstance and given the user’s open input.

To be capable of efficiently transmitting appropriate information on the motion improvement to a person in motion. A motion evaluation system includes a sensor unit, an information processing device, and an information presentation device. The information processing device includes a communication device, a storage device, and an arithmetic device. The arithmetic device acquires motion data acquired by observing a user through the use of a sensor via the communication device, checks the motion data against information about the correctness of motions in the reference information, determines a state of motion of the user, specifies a motion in a state to be improved as an improvement, check the motion data after the motion corresponding to the improvement against information about busy levels of the user to specify a busy level of the user, and outputs, as improvement suggestion information about the improvement, information with different contents at each of multiple times to an information presentation device based on the improvement and a rule predetermined according to each situation of the busy level.

The present disclosure relates to technology that controls a remote moving body based on collaboration between the moving body and human, and a method for controlling a moving body includes acquiring a first biosignal indicating an intention to start operation of the moving body from a user, operating the moving body, determining a surrounding situation of the moving body that autonomously controls the driving, providing the user with surrounding information of the moving body for inducing path setting, acquiring a second biosignal evoked by recognition of the surrounding information from the user, setting a driving direction of the moving body, commanding the moving body to automatically perform a driving operation to be carried out in the set driving direction, and acquiring a third biosignal responsive to recognition of a driving error from the user and correcting the driving direction of the moving body to induce driving path resetting.

Systems and methods for detecting, decoupling and quantifying unresolved signals in trace signal data in the presence of noise with no prior knowledge of the signal characteristics (e.g., signal peak location, intensity and width) of the unresolved signals. The systems and methods are useful for analyzing any trace data signals having one or multiple overlapping constituent signals and particularly useful for analyzing data signals which often contain an unknown number of constituent signals with varying signal characteristics, such as peak location, peak intensity and peak width, and varying resolutions. A general signal model function is assumed for each unknown, constituent signal in the trace signal data. In a first phase, the number of constituent signals and signal characteristics are determined automatically in a parallel fashion by executing multiple simultaneous evaluations iteratively starting with an initial set of trial signals. Making simultaneous evaluations and systematically reducing the number of trial signals allows for convergence to an optimal, final set of signals in a very fast and efficient manner.

There is provided an information processing apparatus to detect various motions of a user and generate useful information utilizing the result, the information processing apparatus including: a motion detection unit configured to detect a series of motions of a user which are repetitive and nonperiodic; and an information generation unit configured to generate information related to the series of motions.