ACTION SERIES DETERMINATION DEVICE, METHOD, AND NON-TRANSITORY RECORDING MEDIUM

    公开(公告)号:US20230377374A1

    公开(公告)日:2023-11-23

    申请号:US18341566

    申请日:2023-06-26

    Abstract: From plural observation features of a time series acquired by observing movements of a person, plural candidate segments are decided of a target action series containing plural respective actions expressing plural movements. Each of the plural candidate segments is divided into each action segment that is a time segment of the action, a likelihood corresponding to each of the plural actions computed for each of the action segments is normalized by action segment, and as an evaluation value a representative value is computed of normalized likelihood corresponding to each of the action segments selected from out of all of the respective action segments in the candidate segments based on an order of actions in the target action series. Being the target action series is determined in cases in which the evaluation value exceeds a common threshold.

    COMPUTER-READABLE RECORDING MEDIUM STORING INFORMATION PROCESSING PROGRAM, INFORMATION PROCESSING METHOD, AND INFORMATION PROCESSING APPARATUS

    公开(公告)号:US20240428593A1

    公开(公告)日:2024-12-26

    申请号:US18667419

    申请日:2024-05-17

    Inventor: Junya FUJIMOTO

    Abstract: A recording medium storing a program for causing a computer to execute processing including: acquiring a video data including frames each of which is associated with a time point and includes the video captured at the time point, by capturing a video at a site in which a person and an object exist; identifying a first relation by analyzing a first frame associated with a first time point, the first relation identifying an interaction between the object and the person in the first frame; identifying a second relation by analyzing a second frame associated with a second time point, the second relation identifying an interaction between the object and the person in the second frame; and specifying, based on the first and second relations, an action of the person on the object in a third frame associated with a third time point between the first and second time points.

    COMPUTER-READABLE RECORDING MEDIUM STORING INFORMATION PROCESSING PROGRAM, INFORMATION PROCESSING METHOD, AND INFORMATION PROCESSING APPARATUS

    公开(公告)号:US20250069357A1

    公开(公告)日:2025-02-27

    申请号:US18942868

    申请日:2024-11-11

    Abstract: A non-transitory computer-readable recording medium storing a program for causing a computer to execute a process includes generating three-dimensional first skeleton information of a first person included in a first video by performing skeleton recognition, setting a first interesting region in a three-dimensional space for each of objects included in the first video based on design data, specifying a first object being used by the first person from among the objects included in the first video by using positional information of a hand of the first skeleton information and the first region of interest, acquiring a position distribution in which the hand is present in the three-dimensional space based on a trajectory of the part of the hand of the first skeleton information, and setting a second region of interest, in the three-dimensional space, that indicates that a person uses the first object, based on the position distribution.

    PARTIAL ACTION SEGMENT ESTIMATION MODEL BUILDING DEVICE, METHOD, AND NON-TRANSITORY RECORDING MEDIUM

    公开(公告)号:US20230343080A1

    公开(公告)日:2023-10-26

    申请号:US18341548

    申请日:2023-06-26

    CPC classification number: G06V10/7788 G06V40/23 G06V10/85

    Abstract: A hidden semi-Markov model includes plural second hidden Markov models each containing plural first hidden Markov models using types of movement of a person as states. The plural second hidden Markov models each use partial actions that are parts of actions determined by combining plural movements as states. In the hidden semi-Markov model observation probabilities are leant for each type of the movements of the plural first hidden Markov models using unsupervised learning. The learnt observation probabilities are fixed, and input first supervised data is augmented to give second supervised data, and transition probabilities of the movements of the first hidden Markov models are learned by supervised learning in which the second supervised data is employed. The learnt observation probabilities and transition probabilities are employed to build the hidden semi-Markov model that is a model for estimating segments of the partial actions.

Patent Agency Ranking