Panasonic Intellectual Property Management Co., Ltd.

Ariel Beck

Chandra Suwandi Wijaya

Khai Jun Kek

Vasileios Vonikakis

Xibeijia Guan

Nway Nway Aung

Yao Zhou

Athul M Mathew

Zong Sheng Tang

Yuya Sugasawa

Souksakhone Bounyong

Eng Chye Lim

Pradeep Rajagopalan

Jeevan Kumar Guntammagari

Sahim Yamaura

Muhammad Usman

Myo Htun

Andre Ivan

Myo Min Latt

Roy Eng Chye Lim

Ramdas Krishnakumar

The present disclosure relates to a method for estimating physical properties of an object. The method includes generating a three-dimensional (3D) reconstruction of a real-world environment based on an input media. Further, the method includes estimating a current set of physical parameters of the object using a physics-based prediction model and a real-world visual content while the object is manipulated. Further, the method includes generating a simulation variant and computing a cost function based on comparing the simulation variant and the real-world visual content. Furthermore, the method includes estimating the physical properties of the object based on optimizing the current set of physical parameters with the cost function such that a similarity between the simulation variant and the real-world visual content increases.

Methods and systems for estimating physical properties of objects

The present subject matter refers a method for developing machine-learning (ML) based tool. The method comprises initializing an input dataset for undergoing ML based processing. The input dataset is pre-processed by a first model to harmonize features across the dataset. Thereafter, the dataset is annotated by a second model to define a labelled data set. A plurality of features are extracted with respect to the data set through a feature extractor. A selection of at-least a machine-learning classifier is received through an ML training module to operate upon the extracted features and classify the dataset with respect to one or more labels. A meta controller communicated with one or more of the first model, the second model, the feature extractor and the selected classifier for assessing a performance of at least one of first model and the feature extractor, a comparison of operation among the one or more selected classifier, and diagnosis of an unexpected operation with respect to one or more of the first model, the feature extractor and the selected classifier.

Method for developing machine-learning based tool

The present subject matter discloses a system(s) and a method(s) for determining a health state of an individual. According to an embodiment, a method comprises measuring, by a heart rate sensor, a heart rate of the individual during operation within the environment. The method further comprises outputting, by a pressure sensing platform, pressure data of the individual. Further, the method comprises outputting, by an image capturing device, image data of the individual. The method further comprises inferring, by a processing unit, an amount of fat of the individual in the image data. The method further comprises updating, by the processing unit, the amount of fat of the individual using the pressure data. The method further comprises controlling, by the processing unit, a threshold for determining the health state of the individual, using the amount of fat and the heart rate of the individual.

Determining health state of individuals

A method for developing machine-learning (ML) based tool including initializing an input dataset, which is pre-processed by a first model to harmonize the dataset. Historical data similar to the input data set is fetched from a historical database. Based thereupon a controller recommends a method and a control-setting associated with the identified model for the visual inspection process to a user. Thereafter, the dataset is annotated by a second model to define a labelled data set. A plurality of features are extracted with respect to the data set through a feature extractor. A machine-learning classifier operates upon the extracted features and classifies the dataset with respect to one or more labels. A meta controller communicates with one or more of the first model, the second model, the feature extractor and the selected classifier for assessing a performance of at least one of first model and the feature extractor.

A method and system for checking data gathering conditions or image capturing conditions associated with images during AI based visual-inspection process. The method comprises generating a first representative (FR) image for a first group of images and a second representative image (FR) for a second group of images. A difference image data is generated between FRimage and the FRimage based on calculating difference between luminance values of pixels with same coordinate values. Thereafter, one or more of a plurality of white pixels or intensity-values are determined within the difference image based on acquiring difference image data formed of luminance difference-values of pixels. An index representing difference of data-capturing conditions across the FRimage and the FRimage is determined, said index having been determined at least based on the plurality of white pixels or intensity-values, for example, based on application of a plurality of AI or ML techniques.

Method and system for checking data gathering conditions associated with image-data during AI enabled visual-inspection process

A graphical user interface (GUI) for forming hierarchically arranged clusters of items and operating thereupon through an electronic device equipped with an input-device and a display-screen is provided. The GUI comprises a first area configured to display a graphical-tree representation having a plurality of hierarchical levels, each of said level corresponds to at least one cluster of content-items formed by execution of a machine-learning classifier over a plurality of input content items. A second area is configured to display a dataset corresponding to the content-items classified within the clusters. A third area is configured to display a plurality of types of content representations with respect to each selected cluster, said representations corresponding to content-items classified within the cluster.

Methods and systems for monitoring objects for labelling

The present subject matter refers a method for training image-alignment procedures in a computing environment. The method comprises communicating one or more images of an object to a user and receiving a plurality of user-selected zones within said one or more through a user-interface. An augmented data-set is generated based on said one or more images comprising the user-selected zones, wherein such augmented data set comprises a plurality of additional images defining variants of said one or more communicated images. Thereafter, a machine-learning based image alignment is trained based on at-least one of the augmented data set and the communicated images.

Method and system for training image-alignment procedures in computing environment

The present disclosure relates to a method and system for apportioning tasks to person in an environment. The method comprises capturing a first-value indicating a sympathetic-nerve based activity and a second-value indicating a parasympathetic-nerve based activity for at least one person operating in an environment. Thereafter, a quantitative-relation is determined between the first and second values. At-least one task is assigned for execution by said person within the environment based on such quantitative relation.

Method and system for apportioning tasks to persons in environment

A system for gaze direction detection is disclosed herein. The system comprises a face detection unit configured to detect face information from a captured image of a spectator, a head pose unit configured to detect a head pose angle based on the face information; an eye measurement unit configured to detect eye information of at-least one of a left eye or a right eye of the spectator based on the detected face information; an eye-gaze detection unit configured to estimate a gaze angle of the at-least one of the left eye or the right eye of the subject based on the eye information; a conditional gaze detection unit configured to calculate a resultant gaze angle based on the estimated gaze angles of the left eye and right eye and the head pose angle; and an effective gaze detection unit configured to estimate an effective gaze angle based the resultant gaze angle and head-pose angle.

System and method for gaze direction detection

A method implemented in a computing-device with a display screen for image inspection. The method comprises displaying a distribution of a quality-indicia of at least one object in each of a plurality of images to be inspected, within a first area of the display screen. Within a second area of the display screen, a user-control is displayed to adjust a threshold-value with respect to an acceptance of at least one object in said plurality of images to be inspected. The threshold-value may be determined manually or automatically. A change in or update of threshold value is determined based on a user-operation performed over the user-control for adjusting the threshold value. Thereafter, a quality-indicia of at least one object in each the plurality of images is determined. Acceptable objects in respect of an image inspection procedure based on the updated threshold value and the determined quality-indicia are indicated.

Methods and systems for monitoring objects for image-inspection

The present subject matter refers an image-processing method comprises receiving a first-image of an object captured by a range-imaging device at a first viewing location. The transforming the first-image into a second image of the object, said second image corresponding to an image captured based on range-imaging at a second viewing location with respect to the object. The gaps in the second-image are identified based on comparison with the first image, such that the identified gaps within the second image are complemented to result in a complemented second image.

Image-processing method

There is provided a computer-implemented method for generating an output with respect to a group of individuals. The method includes: identifying the group of individuals amongst a plurality of individuals in an area being monitored by one or more sensors; determining, for each individual in the group of individuals, one or more individual-based features associated with the individual based on sensing data obtained from the one or more sensors; determining a group characteristic information associated with the group of individuals based on the one or more individual-based features determined for each individual in the group; and generating the output based on the group characteristic information determined for the group of individuals. There is also provide a corresponding system for generating an output with respect to a group of individuals.

Method and system for generating an output with respect to a group of individuals

A system and method for monitoring the condition of the driver of a vehicle including both the level of distraction being experienced by him/her as well his/her cardiac health are described. The system can detect and predict behavioral and cognitive distraction in real time. It can also monitor a driver's physical condition, detect and predict health events such as a myocardial infarction. The system can fuse inputs from various sensors and components including vehicle information (through CAN networks or other), IR camera, heart rate sensors, eye tracking and pupillometry technology. The distraction detection and health monitoring can be done in separate modules. Fusion algorithms can detect and predict risk related to the driver's lack of attention and the likelihood that he/she is or will experience a cardiac event.

Growing community of inventors

Singapore, Singapore

Ariel Beck