知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

42 records (took 0.018 seconds)

    Augmenting virtual reality content with real world content
    12.
    发明授权
    Augmenting virtual reality content with real world content 有权

    公开(公告)号：US10191541B2

    公开(公告)日：2019-01-29

    申请号：US15385794

    申请日：2016-12-20

    Applicant: Sony Interactive Entertainment Inc.

    Inventor： Ruxin Chen

    IPC: G06F3/01 ,  G02B27/01 ,  G06T11/60 ,  G09G5/36 ,  G06F3/03 ,  G06F3/0481 ,  G06F3/0346

    Abstract: Methods, devices, and computer programs for augmenting a virtual reality scene with real world content are provided. One example method includes an operation for obtaining sensor data from an HMD of a user to determine that a criteria is met to overlay one or more real world objects into the virtual reality scene to provide an augmented virtual reality scene. In certain examples, the criteria corresponds to predetermined indicators suggestive of disorientation of a user when wearing the HMD and being presented a virtual reality scene. In certain other examples, the one or more real world objects are selected based on their effectiveness at reorienting a disoriented user.

    Deep reinforcement learning framework for characterizing video content
    18.
    发明授权
    Deep reinforcement learning framework for characterizing video content 有权

    公开(公告)号：US10885341B2

    公开(公告)日：2021-01-05

    申请号：US16171018

    申请日：2018-10-25

    Applicant: Sony Interactive Entertainment Inc.

    Inventor： Ruxin Chen ,  Naveen Kumar ,  Haoqi Li

    IPC: G06K9/00 ,  G06N20/00 ,  G06K9/62 ,  G06N3/08 ,  G06N3/00

    Abstract: Methods and systems for performing sequence level prediction of a video scene are described. Video information in a video scene is represented as a sequence of features depicted each frame. An environment state for each time step t corresponding to each frame is represented by the video information for time step t and predicted affective information from a previous time step t−1. An action A(t) as taken with an agent controlled by a machine learning algorithm for the frame at step t, wherein an output of the action A(t) represents affective label prediction for the frame at the time step t. A pool of predicted actions is transformed to a predicted affective history at a next time step t+1. The predictive affective history is included as part of the environment state for the next time step t+1. A reward R is generated on predicted actions up to the current time step t, by comparing them against corresponding annotated movie scene affective labels.

    System and method for converting image data into a natural language description
    19.
    发明授权
    System and method for converting image data into a natural language description 有权

    公开(公告)号：US10726062B2

    公开(公告)日：2020-07-28

    申请号：US16206439

    申请日：2018-11-30

    Applicant: Sony Interactive Entertainment Inc.

    Inventor： Jian Zheng ,  Ruxin Chen

    IPC: G06F16/383 ,  G06F16/583 ,  G06K9/32 ,  G06N5/04

    Abstract: For image captioning such as for computer game images or other images, bottom-up attention is combined with top-down attention to provide a multi-level residual attention-based image captioning model. A residual attention mechanism is first applied in the Faster R-CNN network to learn better feature representations for each region by taking spatial information into consideration. In the image captioning network, taking the extracted regional features as input, a second residual attention network is implemented to fuse the regional features attentionally for subsequent caption generation.

    Initialization of CTC speech recognition with standard HMM
    20.
    发明授权
    Initialization of CTC speech recognition with standard HMM 有权

    公开(公告)号：US10714076B2

    公开(公告)日：2020-07-14

    申请号：US15645985

    申请日：2017-07-10

    Applicant: Sony Interactive Entertainment Inc.

    Inventor： Xavier Menendez-Pidal ,  Ruxin Chen

    IPC: G10L15/00 ,  G10L15/16 ,  G10L15/14 ,  G10L15/02 ,  G10L15/06

    Abstract: A method for improved initialization of speech recognition system comprises mapping a trained hidden markov model based recognition node network (HMM) to a Connectionist Temporal Classification (CTC) based node label scheme. The central state of each frame in the HMM are mapped to CTC-labeled output nodes and the non-central states of each frame are mapped to CTC-blank nodes to generate a CTC-labeled HMM and each central state represents a phoneme from human speech detected and extracted by a computing device. Next the CTC-labeled HMM is trained using a cost function, wherein the cost function is not part of a CTC cost function. Finally the CTC-labeled HMM is trained using a CTC cost function to produce a CTC node network. The CTC node network may be iteratively trained by repeating the initialization steps.

Patent Agency Ranking