利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

204 条记录 (耗时 0.016 秒)

    文字候補提案装置、手書文字識別システム、手書文字識別方法およびプログラム
    2.
    发明申请
    文字候補提案装置、手書文字識別システム、手書文字識別方法およびプログラム 审中-公开

    公开(公告)号：WO2021049070A1

    公开(公告)日：2021-03-18

    申请号：PCT/JP2020/011553

    申请日：2020-03-16

    申请人： 株式会社日立システムズ

    发明人： 北見　淳

    IPC分类号： G06K9/68 ,  G06K9/72

    摘要： 類似する文字列候補をより適切に提案する技術を提供する。　文字候補提案装置であって、文字の外観の類似関係を記憶する類似文字記憶部と、実存する文字列を記憶する実存文字記憶部と、一つ以上の文字を受信して該文字の外観の類似する文字候補の案出を受託する受託処理部と、受信した文字と外観が類似する文字の候補を、類似文字記憶部と照合することで特定し、外観が類似する文字の候補の組み合わせのうち実存文字記憶部に記憶された実存する文字列と照合する照合処理部と、照合処理部による照合の結果ヒットした外観が類似する文字の候補の組み合わせを候補文字として出力する候補文字送信処理部と、を備える。

    SURFACE PRESENTATIONS
    3.
    发明申请
    SURFACE PRESENTATIONS 审中-公开

    公开(公告)号：WO2021021154A1

    公开(公告)日：2021-02-04

    申请号：PCT/US2019/044324

    申请日：2019-07-31

    申请人： HEWLETT-PACKARD DEVELOPMENT COMPANY, L.P.

    发明人： PARK, Sook Min

    IPC分类号： H04N7/15 ,  G06K9/72

    摘要： Examples of computing devices are described herein. In some examples, a computing device includes machine-readable instructions stored in a non-transitory storage medium. In some examples, the instructions are executable by a processor to determine a content feed using artificial intelligence based on a captured image of a writing surface. In some examples, the instructions are executable to present the content feed with a representation of the writing surface.

    SELECTING UNLABELED DATA OBJECTS TO BE PROCESSED
    4.
    发明申请
    SELECTING UNLABELED DATA OBJECTS TO BE PROCESSED 审中-公开

    公开(公告)号：WO2020014778A1

    公开(公告)日：2020-01-23

    申请号：PCT/CA2019/050978

    申请日：2019-07-16

    申请人： ELEMENT AI INC.

    发明人： ROBERT, Eric ,  BÉJEAU, Jean-Sébastien

    IPC分类号： G06K9/72 ,  G06N3/08 ,  G06F16/906

    摘要： Systems and methods for selecting at least one unlabeled data object from a set of unlabeled data objects. The present invention receives a set of unlabeled data objects and identifies at least one data object in the set that is considered to differ from the others. The at least one data object is selected for further processing, which may include labeling processes. In some embodiments, the data objects are passed through at least one representation-generating module, and the resulting representations are compared to each other. Differences between the representations are evaluated against at least one criterion. If the differences meet the at least one criterion, corresponding data objects are considered to differ from the others and are then selected for further processing. In some implementations, a sample set of sample data objects may be used. In some implementations, the at least one representation-generating module may comprise a neural network.

    INTERACTIVE AND ADAPTIVE LEARNING, NEUROCOGNITIVE DISORDER DIAGNOSIS, AND NONCOMPLIANCE DETECTION SYSTEMS USING PUPILLARY RESPONSE AND FACE TRACKING AND EMOTION DETECTION WITH ASSOCIATED METHODS
    6.
    发明申请
    INTERACTIVE AND ADAPTIVE LEARNING, NEUROCOGNITIVE DISORDER DIAGNOSIS, AND NONCOMPLIANCE DETECTION SYSTEMS USING PUPILLARY RESPONSE AND FACE TRACKING AND EMOTION DETECTION WITH ASSOCIATED METHODS 审中-公开

    公开(公告)号：WO2019111165A1

    公开(公告)日：2019-06-13

    申请号：PCT/IB2018/059653

    申请日：2018-12-05

    申请人： LAM, Yuen Lee Viola

    发明人： LAM, Yuen Lee Viola

    IPC分类号： G06K9/72

    摘要： A system for detection of noncompliance including substance abuse, driving under influence, and untruthful testimony giving under law enforcement setting, comprising optical sensors for capturing a subject's pupillary responses, blinking eye movements, point-of-gaze, facial expression, and head pose during a compliance test session. The system can also be applied in neurocognitive disorder diagnosis. The subject's affective and cognitive states estimation based on the captured sensory data during a diagnosis test is feedback to the system to drive the course of the compliance or cognitive test, adaptively change the test materials, and influence the subject's affective and cognitive states. The estimated affective and cognitive states in turn provide a more accurate reading of the subject's condition.

    TRACKING A HEAD OF A SUBJECT
    8.
    发明申请
    TRACKING A HEAD OF A SUBJECT 审中-公开

    公开(公告)号：WO2018108926A1

    公开(公告)日：2018-06-21

    申请号：PCT/EP2017/082453

    申请日：2017-12-12

    申请人： KONINKLIJKE PHILIPS N.V.

    发明人： VAN BREE, Karl, Catharina ,  VELTHOVEN, Leo, Jan

    IPC分类号： G06T7/20 ,  G06K9/72 ,  G06K9/00

    摘要： According to an aspect, there is provided a method for tracking a head of a subject in a series of images. The method comprises receiving a series of images of at least a portion of a subject's head; tracking a position of the subject's head in the series of images based on positions of a first plurality of primary features of the subject's head; determining a first confidence level of the position of the subject's head based on the positions of the first plurality of primary features; monitoring a position of a secondary feature; and upon determining that the first confidence level is below a defined threshold, tracking the position of the subject's head based on the position of the secondary feature rather than the positions of the primary features. According to other aspects, a processing unit, an apparatus and a machine readable medium are provided.

    SYSTEM AND METHOD FOR SENTENCE DIRECTED VIDEO OBJECT CODETECTION
    9.
    发明申请
    SYSTEM AND METHOD FOR SENTENCE DIRECTED VIDEO OBJECT CODETECTION 审中-公开
    系统和方法的句子导向视频对象编码

    公开(公告)号：WO2017214208A1

    公开(公告)日：2017-12-14

    申请号：PCT/US2017/036232

    申请日：2017-06-06

    申请人： PURDUE RESEARCH FOUNDATION

    发明人： SISKIND, Jeffrey, Mark ,  YU, Haonan

    IPC分类号： G06F17/30 ,  G06K9/62 ,  G06K9/66 ,  G06K9/72

    摘要： A system and method for determining the locations and types of objects in a plurality of videos. The method comprises pairing each video with one or more sentences describing the activity or activities in which those objects participate in the associated video, wherein no use is made of a pretrained object detector. The object locations are specified as rectangles, the object types are specified as nouns, and sentences describe the relative positions and motions of the objects in the videos referred to by the nouns in the sentences. The relative positions and motions of the objects in the video are described by a conjunction of predicates constructed to represent the activity described by the sentences associated with the videos.

    摘要翻译： 用于确定多个视频中的对象的位置和类型的系统和方法。 该方法包括将每个视频与描述这些对象参与相关视频的活动或活动的一个或多个句子配对，其中不使用预训练对象检测器。 对象位置被指定为矩形，对象类型被指定为名词，并且句子描述句子中名词所指的视频中的对象的相对位置和运动。 视频中对象的相对位置和运动由谓词的连接词来描述，这些谓词被构造来表示由与视频相关的句子描述的活动。

    認識装置、認識方法およびプログラム
    10.
    发明申请
    認識装置、認識方法およびプログラム 审中-公开
    识别装置，识别方法和程序

    公开(公告)号：WO2016181458A1

    公开(公告)日：2016-11-17

    申请号：PCT/JP2015/063482

    申请日：2015-05-11

    申请人： 株式会社東芝 ,  東芝ソリューション株式会社

    发明人： 鈴木　智久

    IPC分类号： G06K9/72

    CPC分类号： G06K9/723 ,  G06K9/6204 ,  G06K9/6277 ,  G06K9/72 ,  G06K2209/011

    摘要： 　照合処理に同期させて照合処理とは異なる処理を実行する。入力画像から、文字を含むと推測される画素の集合である文字候補を検出する候補検出部と、文字候補のそれぞれを認識して、認識結果の候補の文字である少なくとも１つの認識候補を生成する認識部と、少なくとも１つの認識候補のそれぞれを、認識対象の文字列をモデル化した知識辞書と照合して、入力画像に含まれると推測される文字列と知識辞書を照合して得られる照合結果を生成する照合部と、与えられた命令を実行する命令処理部を備え、知識辞書は、有限オートマトンであって文字を表す実文字コードおよび命令を指定する仮想文字コードのいずれか一方がエッジに割り当てられ、照合部は、仮想文字コードが割り当てられたエッジに従って知識辞書の状態を遷移させる場合、当該エッジに割り当てられた仮想文字コードにより指定された命令を命令処理部に与える。

    摘要翻译： 为了与匹配处理同步地执行匹配处理以外的处理，该识别装置具备：候选检测单元，其从输入图像检测作为估计为包含字符的像素组的字符候选; 识别单元，其对每个所述字符候选进行识别，并且生成作为由所述识别产生的候选字符的至少一个识别候选; 匹配单元，其将所述至少一个识别候选中的每一个与知识词典相匹配，知识词典包括用于识别的字符串的模型，并且通过将所估计的包含在输入图像中的字符串与知识词典进行匹配来生成匹配结果; 以及执行给出的命令的命令处理单元。 在作为有限自动机的知识词典中，表示字符的实际字符代码或指示命令的虚拟字符代码被分配给边缘; 在根据已经分配了虚拟字符代码的边缘转换知识词典的状态的情况下，匹配单元给予命令处理单元由分配给所述边缘的虚拟字符代码指示的命令。