INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD, PROGRAM, AND RECORDING MEDIUM
    5.
    发明申请
    INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD, PROGRAM, AND RECORDING MEDIUM 审中-公开
    信息处理装置,信息处理方法,程序和记录介质

    公开(公告)号:US20120328208A1

    公开(公告)日:2012-12-27

    申请号:US13523538

    申请日:2012-06-14

    IPC分类号: G06K9/36

    摘要: Provided is an information processing apparatus including a low-bit basic image generating part for generating a low-bit basic image having a low bit depth, a low-bit reference image generating part for generating a low-bit reference image having a low bit depth, a feature value calculating part for calculating a feature value indicating a non-flatness of brightness information of the basic image, a cost value calculating part for calculating a cost value at each of reference block positions in a block matching between the low-bit basic image and the low-bit reference image using the feature value and an estimated motion vector, and a block matching part for calculating a motion vector at each of the reference block positions based on an evaluation value obtained by the block matching between the low-bit basic image and the low-bit reference image after correcting the evaluation value using the calculated cost value.

    摘要翻译: 提供了一种信息处理设备,包括用于产生具有低位深度的低位基本图像的低位基本图像生成部件,用于产生具有低位深度的低位参考图像的低位参考图像生成部件 ,用于计算表示基本图像的亮度信息的非平坦度的特征量的特征量计算部,成本值计算部,用于计算低比特基础的块匹配中的每个参考块位置处的成本值 图像和使用特征值和估计的运动矢量的低位参考图像,以及块匹配部分,用于基于通过低位比特之间的块匹配获得的评估值来计算每个参考块位置处的运动矢量 基本图像和使用计算出的成本值校正评估值之后的低位参考图像。

    METHODS AND SYSTEMS FOR SUPPORT POLICY LEARNING

    公开(公告)号:US20210357782A1

    公开(公告)日:2021-11-18

    申请号:US16875741

    申请日:2020-05-15

    IPC分类号: G06N5/04 G06N20/00 G06N3/00

    摘要: Methods and systems are described for support policy learning in an agent of a robot. A general value function (GVF) is learned for a main policy, where the GVF represents future performance of the agent executing the main policy for a given state of the environment. A master policy selects an action based on the predicted accumulated success value received from the general value function. When the predicted accumulated success value is an acceptable value, the action selected by the master policy is execution of the main policy. When the predicted accumulated success value is not an acceptable value, the master action causes a support policy to be learned. The support policy generates a support action to be performed which causes the robot to transition from to a new state where the predicted accumulated success value has an acceptable value.

    Information Processing Device, Information Processing Method, and Program
    10.
    发明申请
    Information Processing Device, Information Processing Method, and Program 审中-公开
    信息处理装置,信息处理方法和程序

    公开(公告)号:US20110194607A1

    公开(公告)日:2011-08-11

    申请号:US12980946

    申请日:2010-12-29

    IPC分类号: H04N7/26

    CPC分类号: H04N19/53

    摘要: An information processing device includes a bit conversion unit that performs reduction in a bit number allocated to a pixel for a criterion image and a reference image and generates a criterion image and a reference image which are hierarchized according to a bit number allocated to a pixel, and a motion vector detection unit that performs coarse detection of a motion vector by block matching using the criterion image and the reference image of which a bit number is reduced, decides a search range for fine detection of a motion vector based on the motion vector detected by the coarse detection, and performs fine detection of a motion vector using an image in the search range which belongs to a hierarchy equal to or more than an image used in the coarse detection in a bit number allocated to a pixel.

    摘要翻译: 一种信息处理装置,包括:位转换单元,其对分配给标准图像的像素和参考图像进行比特数的减少,并且生成根据分配给像素的比特数分级的标准图像和参考图像, 以及运动矢量检测单元,其使用标准图像和比特数减少的参考图像通过块匹配来执行运动矢量的粗略检测,基于检测到的运动矢量来确定运动矢量的精细检测的搜索范围 通过粗略检测,并且使用属于等于或大于分配给像素的比特数中的粗略检测中使用的图像的等级的搜索范围内的图像来执行运动矢量的精细检测。