利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

244 条记录 (耗时 0.018 秒)

    VIDEO RECOMMENDATION BASED ON VIDEO TITLES
    2.
    发明公开
    VIDEO RECOMMENDATION BASED ON VIDEO TITLES 审中-公开

    公开(公告)号：EP3341861A1

    公开(公告)日：2018-07-04

    申请号：EP16838627.4

    申请日：2016-08-09

    申请人： Google LLC

    发明人： CHECKLEY, Garen

    IPC分类号： G06F17/30 ,  H04N21/83

    CPC分类号： G06F17/3053 ,  G06F17/3082 ,  G06F17/30867

    摘要： A solution is provided to generate video recommendations in a video sharing environment. A video recommendation system selects a video as a target video and extracts target keywords from the title of the identified target video or the title of a non-video trending news item. The system receives multiple candidate videos. For each candidate video, the system extracts keywords from the title of the candidate video and compares the extracted keywords with the target words. Based on the comparison, the system generates a similarity score for the candidate video. The system ranks the candidate videos based their associated similarity scores and selects a candidate video having the highest similarity score as the video recommendation for the target video.

    VIDEO SEARCH SYSTEM & METHOD
    3.
    发明公开
    VIDEO SEARCH SYSTEM & METHOD 审中-公开
    视频搜索系统和方法

    公开(公告)号：EP3301596A1

    公开(公告)日：2018-04-04

    申请号：EP17275153.9

    申请日：2017-09-29

    申请人： British Broadcasting Corporation

    发明人： EGGINK, Jana ,  CLARK, Richard

    IPC分类号： G06F17/30

    CPC分类号： G06F17/3082 ,  G06F17/30784 ,  G06K9/00758 ,  G06K9/00765

    摘要： A system for identifying one or more locations of a video clip within video content comprises means for producing segments of the video clip each segment derived from a plurality of frames. The segments are compared to segments of video content. From the comparison a first measure of similarity is prepared at virtually displaced temporal positions of the video clip and the video content. A filtering arrangement filters the measure of similarity to exclude potential matches based on temporal separation of positions of high similarity. A means is then arranged to compare the video clip to the video content at the positions identified as candidate matches. In this way, apparent matches can be removed based on likely knowledge of the content such as likely repetitions of a video clip within larger video content.

    摘要翻译： 一种用于识别视频内容内的视频剪辑的一个或多个位置的系统包括用于产生视频剪辑的片段的装置，每个片段从多个帧导出。 这些片段与视频内容片段进行比较。 根据比较，在视频剪辑和视频内容的虚拟移位的时间位置处准备第一类似度度量。 过滤装置过滤相似性度量以排除基于高相似度位置的时间分离的潜在匹配。 然后安排一种手段将视频剪辑与识别为候选匹配的位置处的视频内容进行比较。 以这种方式，可以基于内容的可能知识去除明显的匹配，例如在较大的视频内容中可能重复的视频剪辑。

    GENERATING RECOMMENDATIONS BASED ON PROCESSING CONTENT ITEM METADATA TAGS
    4.
    发明公开
    GENERATING RECOMMENDATIONS BASED ON PROCESSING CONTENT ITEM METADATA TAGS 审中-公开
    基于处理内容项目元数据标签生成建议

    公开(公告)号：EP3230937A1

    公开(公告)日：2017-10-18

    申请号：EP15816944.1

    申请日：2015-11-30

    申请人： Microsoft Technology Licensing, LLC

    发明人： MCCLANAHAN, Joseph Lloyd ,  KULKARNI, Asmita ,  SEIN, Travis Albert

    IPC分类号： G06Q30/02

    CPC分类号： H04N21/26208 ,  G06F17/30044 ,  G06F17/30598 ,  G06F17/30752 ,  G06F17/3082 ,  G06Q30/02 ,  G06Q30/0251 ,  G06Q30/0255 ,  G06Q30/0269 ,  H04N21/251 ,  H04N21/25891 ,  H04N21/26258 ,  H04N21/2668 ,  H04N21/44222 ,  H04N21/812

    摘要： A device, method, and computer-readable media for managing interrupt times for content items based on metadata encapsulating user behavior. A user controls the playback of content items such as audio or videos. The content items may be episodes in a programming series. Tags are associated with the content items. These tags include metadata having playback time of the received user interactions. In turn, the device processes the tags to identify user interaction that include stop events for the content items. The tags are processed based on a sliding time window. The playback times of the stop events are selected as potential interrupt times that the device may include recommendations for unaired episodes. The selected interrupt times are also used to identify inconsistencies in the metadata.

    摘要翻译： 一种用于基于封装用户行为的元数据来管理内容项的中断时间的设备，方法和计算机可读介质。 用户控制诸如音频或视频的内容项目的回放。 内容项目可以是节目系列中的剧集。 标签与内容项目相关联。 这些标签包括具有接收到的用户交互的回放时间的元数据。 接着，设备处理标签以识别包括内容项的停止事件的用户交互。 标签是基于滑动时间窗口进行处理的。 停止事件的回放时间被选择为潜在的中断时间，该设备可以包括针对不期望的情节的推荐。 选定的中断时间也用于识别元数据中的不一致性。

    VIDEO CATEGORIZATION METHOD AND APPARATUS, COMPUTER PROGRAM AND RECORDING MEDIUM
    7.
    发明公开
    VIDEO CATEGORIZATION METHOD AND APPARATUS, COMPUTER PROGRAM AND RECORDING MEDIUM 审中-公开
    维多利亚州电力公司，电力公司

    公开(公告)号：EP3176709A1

    公开(公告)日：2017-06-07

    申请号：EP16176268.7

    申请日：2016-06-24

    申请人： Xiaomi Inc.

    发明人： CHEN, Zhijun ,  HOU, Wendi ,  LONG, Fei

    IPC分类号： G06F17/30 ,  G06K9/00

    CPC分类号： G06K9/00718 ,  G06F17/30793 ,  G06F17/3082 ,  G06K9/00268 ,  G06K9/00288 ,  G06K9/4671 ,  G06K9/6215 ,  G06K9/6218 ,  G06K9/6267 ,  G11B27/10 ,  H04N21/8456

    摘要： The present invention relates to a video categorization method and apparatus, a computer program and a recording medium. The method comprises: acquiring a key frame, which contains a face, in a video; acquiring a face feature in the key frame; acquiring one or more face features corresponding to one or more picture categories; determining, according to the face feature in the key frame and the one or more face features corresponding to the one or more picture categories, a picture category to which the video belongs; and assigning the video to the picture category to which the video belongs. With the above technical solution, a video may be smartly and automatically classified into a picture category corresponding to a user appearing in the video, thereby not only eliminating the need for manual categorization by the user but also improving the accuracy of categorization.

    摘要翻译： 本发明涉及视频分类方法和装置，计算机程序和记录介质。 该方法包括：在视频中获取包含脸部的关键帧; 获取关键帧中的人脸特征; 获取与一个或多个图像类别对应的一个或多个面部特征; 根据所述关键帧中的面部特征和与所述一个或多个图像类别相对应的一个或多个面部特征，确定所述视频所属的图像类别; 并将视频分配给视频所属的图片类别。 利用上述技术方案，视频可以被智能地自动分类为与出现在视频中的用户相对应的图片类别，从而不仅消除了用户手动分类的需要，而且提高了分类的准确性。

    ANNOTATION METHOD AND CORRESPONDING DEVICE, COMPUTER PROGRAM PRODUCT AND STORAGE MEDIUM
    8.
    发明公开
    ANNOTATION METHOD AND CORRESPONDING DEVICE, COMPUTER PROGRAM PRODUCT AND STORAGE MEDIUM 审中-公开
    新西兰武装部队的电力公司

    公开(公告)号：EP3162080A1

    公开(公告)日：2017-05-03

    申请号：EP15732606.7

    申请日：2015-06-23

    申请人： Thomson Licensing

    发明人： SIROT, Joël ,  VIGOUROUX, Jean-Ronan ,  CHEVALLIER, Louis

    IPC分类号： H04N21/4725 ,  G06F17/30

    CPC分类号： H04N21/4725 ,  G06F17/30268 ,  G06F17/3079 ,  G06F17/3082 ,  H04N21/41407 ,  H04N21/435 ,  H04N21/4728 ,  H04N21/4788 ,  H04N21/8456 ,  H04N21/8547 ,  H04N21/8583

    摘要： The present disclosure relates to a method for annotating a content element of a video stream which has been at least partially received by an electronic device, said method being implemented by said electronic device during a restitution of said video stream. According to the present disclosure, the method comprises:—receiving at least one item of information for identifying an image part in said video stream, comprising a temporal and/or spatial stamping of said image part;—when said identified image part belongs to a portion already restituted of said video stream: analysing said restituted portion, and obtaining a significant content element from said identified image part; searching for the presence of said significant content element in an image, called marked image, of at least one portion remaining to be restituted of said video stream; when a marked image is found, associating an annotation linked to said content item with a marked image; when no marked image is found, restituting said identified image again, while delivering at least one annotation linked to said content element.

    摘要翻译： 本公开涉及一种用于注释至少部分地由电子设备接收的视频流的内容元素的方法，所述方法由所述电子设备在恢复所述视频流期间实现。 根据本公开，该方法包括： - 接收用于识别所述视频流中的图像部分的至少一个信息项，包括所述图像部分的时间和/或空间冲压; - 当所述识别的图像部分属于 已经恢复所述视频流的部分：分析所述恢复部分，以及从所述识别的图像部分获得重要的内容元素; 搜索被称为标记图像的图像中存在所述有意义的内容元素的至少一部分剩余待恢复的所述视频流; 当找到标记图像时，将链接到所述内容项目的注释与标记图像相关联; 当没有找到标记图像时，在递送链接到所述内容元素的至少一个注释的同时，重新确定所述识别的图像。

    VISUAL ATTENTION DETECTOR AND VISUAL ATTENTION DETECTION METHOD
    9.
    发明公开
    VISUAL ATTENTION DETECTOR AND VISUAL ATTENTION DETECTION METHOD 审中-公开
    VISUELLER AUFMERKSAMKEITSDETEKTOR UND VISUELLES AUFMERKSAMKEITSDETEKTIONSVERFAHREN

    公开(公告)号：EP3151160A1

    公开(公告)日：2017-04-05

    申请号：EP16183153.2

    申请日：2016-08-08

    申请人： OMRON CORPORATION

    发明人： RUAN, Xiang c/o OMRON CORPORATION ,  LU, Huchuan, c/o School of information and

    IPC分类号： G06K9/00 ,  G06K9/46

    CPC分类号： G06K9/4604 ,  G06F17/3082 ,  G06K9/00771 ,  G06K9/4623 ,  G06K9/4642 ,  G06K2009/3291 ,  G06N5/02 ,  G06T7/11

    摘要： To provide a novel algorithm for visual attention detection in videos that can be easily implemented and is of superior reliability, a visual attention detector includes a feature extraction unit configured to extract a spatiotemporal feature from a local region in a video; a hashing unit configured to convert a spatiotemporal feature value for the local region into a hash value, and to select a training value mapped to the hash using a hash table; and an attention measure determining unit for determining an attention measure on the basis of the distance between a spatiotemporal feature value for the local region and the selected training value such that the larger the distance the larger the attention measure.

    摘要翻译： 为了提供可以容易地实现和具有优异可靠性的视频中的视觉注意检测的新颖算法，视觉注意检测器包括被配置为从视频中的局部区域提取时空特征的特征提取单元; 哈希单元，被配置为将本地区域的时空特征值转换成哈希值，并使用哈希表选择映射到哈希的训练值; 以及注意力测定确定单元，用于基于局部区域的时空特征值与所选择的训练值之间的距离来确定注意力测量，使得注意力越大的距离越大。

    METHODS, APPARATUS AND SYSTEMS FOR TIME-BASED AND GEOGRAPHIC NAVIGATION OF VIDEO CONTENT
    10.
    发明公开
    METHODS, APPARATUS AND SYSTEMS FOR TIME-BASED AND GEOGRAPHIC NAVIGATION OF VIDEO CONTENT 审中-公开
    方法，装置和系统基于时间和视频内容的地理导航

    公开(公告)号：EP3123366A1

    公开(公告)日：2017-02-01

    申请号：EP15769452.2

    申请日：2015-03-25

    申请人： TVU Networks Corporation

    发明人： SHEN, Paul ,  MCEWEN, Matthew Richard ,  YAO, Shiwen

    IPC分类号： G06F17/30 ,  G06T1/00

    CPC分类号： G11B27/34 ,  G06F17/30817 ,  G06F17/3082 ,  H04N21/2353 ,  H04N21/2743 ,  H04N21/47205 ,  H04N21/84

    摘要： Methods, apparatus, and systems for time-based and geographic navigation of video content are provided. Video content and associated metadata information are recorded and encoded using a video capture and encoding module. The associated metadata information includes at least one of date and time information of the recording and geographic position information indicative of a recording location. The recorded video content and the associated metadata information are communicated to a remote storage and web server device. A graphical user interface enables the display of an interactive map showing a route and current location of the video capture and encoding module. The video content may be searched using at least one of the graphical user interface and the interactive map by the date and/or time information and the geographic position information. Selected video content can be streamed or downloaded to a select location for display or storage.

    摘要翻译： 方法，设备和系统，用于基于时间和提供的视频内容的地理导航。 视频内容和关联的元数据信息被记录，并使用视频捕捉和编码模块编码。 相关联的元数据信息包括日期和记录和指示记录位置的地理位置信息的时间信息中的至少一个。 记录的视频内容和相关联的元数据信息被传输到远程存储和网服务器装置。 的图形用户界面启用示出视频捕捉和编码模块的路径和当前位置的交互式地图的显示。 视频内容可以使用图形用户界面中的至少一个以及由日期和/或时间信息和所述地理位置信息互动式地图进行搜索。 选择的视频内容可以被流式传输或下载到选择位置显示或存储。