International Business Machines Corporation

Kun Zhao

Takayuki Osogami

Hiroki Nakano

Takayuki Katsuki

Takayuki Yoshizumi

Taro Sekiyama

Tetsuro Morimura

Masaharu Sakamoto

Yasunori Yamada

Yasushi Negishi

Masaharu Sakamato

A computer-implemented method for detecting an operation tendency is disclosed. The method includes preparing a general model for generating a general anomaly score. The method also includes preparing a specific model, for generating a specific anomaly score, trained with a set of a plurality of operation data related to operation by a target operator. The method further includes receiving input operation data. The method includes also calculating a detection score related to the operation tendency by using a general anomaly score and a specific anomaly score generated for the input operation data. Further the method includes outputting a result based on the detection score.

Detection of operation tendency based on anomaly detection

A computer-implemented method of predicting a risk of an accident is disclosed. The method includes computing an anomaly score based on sensor data to obtain a series of anomaly scores. The method also includes processing the anomaly score to limit a processed anomaly score below a predetermined value. The method further includes calculating an accident risk score at time of prediction by using a series of processed anomaly scores up to the time of the prediction. The method includes further outputting a prediction result based on the accident risk score.

Prediction of accident risk based on anomaly detection

Fluid motion is simulated by performing a first fluid simulation without reflecting all of a plurality of forces acting on a fluid, to obtain a first velocity of the fluid at a current time step; estimating a velocity residue at the current time step by inputting a calculated velocity from the previous time step into a regression model. The regression model is trained to relate velocity obtained by performing a second fluid simulation reflecting the plurality of forces acting on the fluid to a difference between the velocity obtained by performing the first fluid simulation and the velocity obtained by performing the second fluid simulation; and calculating a velocity of the fluid at the current time step by adding the first velocity at the current time step and the velocity residue at the current time step.

Machine learning based fluid simulation

A cascading convolutional neural network (CCNN) comprising a plurality of convolutional neural networks (CNNs) that are trained by weighting training data based on loss values of each training datum between CNNs of the CCN. The CCNN can receiving an input image from plurality of images, classify the input image using the CCNN, and present a classification of the input image.

Weighted cascading convolutional neural networks

Methods and systems for game event analysis and control include determining segmentation events in a game that each represent an interruption in game play or a reset scene. An event analysis is performed using the series of events to determine a contribution value for each event in the series of events that represents how much each respective event in the series of events contributed to the point being scored. Events that have a contribution value that exceeds a baseline value for a respective event type are responded to.

Segmentation-based invasion sports analysis and control

A method detects sports highlights for an event involving players and a play object. The method calculates a covariance matrix characterizing a two-dimensional variation of a multivariate probability distribution presumed for the players relative to a spatiotemporal tracking dataset corresponding to the players. The method calculates an area S occupied by the distribution as a function S(t) based on Eigenvalues calculated from the covariance matrix, derives a differential of the function S(t) and obtains an absolute value of the differential using a function &#x192;(t). The method defines a function g(t) relative to a threshold t. The function g(t) assigns a first or second value responsive to a current time t meeting or exceeding, respectively, the threshold t. The method outputs as the sports highlight, a highest value from among local peaks extracted from a convolution of the functions &#x192;(t) and g(t) and being within a range of the threshold t.

Multivariate probability distribution based sports highlight detection

Methods and systems for data visualization include comparing a random value to opacity values along a plurality of rays to determine a stopping point for each ray in a three-dimensional dataset. An expected brightness is determined for each ray based on a brightness value at the stopping point of each ray. An image is generated that visualizes the three-dimensional dataset based on the expected brightness for each ray.

View-dependent stochastic volume rendering with Monte Carlo ray exploration

A computer-implemented method is disclosed. The method includes preparing a base of an anomaly detection model for generating a score that indicates an estimation of a concentration decline. The anomaly detection model has parameters affecting the score. The method also includes preparing a set of training data, each of which includes a sequence of sensor data relating to activity performed by an individual. The method also includes optimizing the parameters of the anomaly detection model using the set of the training data so as to make a score for longer cumulative activity high as compared to shorter cumulative activity. The method further includes outputting the parameters of the anomaly detection model, in which the anomaly detection model having the parameters is used for detecting a concentration decline of a target individual.

Detection of decline in concentration based on anomaly detection

Volume rendering is performed by a method, comprising: obtaining an original volume data, transforming the original volume data based on a distance from a viewpoint to the original volume data, to generate transformed volume data, generating particles from the transformed volume data, and projecting the particles on an image plane to obtain a 2D image corresponding to the original volume data.

Image space-based particle generation modeling

A target object may be identified by estimating a distribution of a plurality of orientations of a periphery of a target object, and identifying the target object based on the distribution.

Growing community of inventors

Funabashi, Japan

Kun Zhao