公开(公告)号：US20080159383A1

公开(公告)日：2008-07-03

申请号：US11717507

申请日：2007-03-12

申请人： Mayank Kukreja ,  Srinivasan H. Sengamedu

发明人： Mayank Kukreja ,  Srinivasan H. Sengamedu

IPC分类号： H04N11/02

CPC分类号： G11B27/34 ,  G06F16/745 ,  G06F16/78 ,  G06F16/784 ,  G06F16/785 ,  G06F16/7857 ,  G06F16/786 ,  G06K9/00711 ,  G11B27/28

摘要： Keyframes of video are arranged on a display based on characteristics on the keyframes, such as content similarity and temporal relation as compared to each other, where input is received comprising one or more keyframes from video data and it is determined where to display the one or more keyframes along a first axis of the display based on a time associated with the keyframe or keyframes. It is determined where to display the one or more keyframes along a second axis based on the content of the keyframe or keyframes.

摘要翻译： 视频的关键帧基于关键帧上的特征被布置在显示器上，例如彼此相比的内容相似性和时间关系，其中接收包括来自视频数据的一个或多个关键帧的输入，并且确定在哪里显示一个或 基于与关键帧或关键帧相关联的时间沿着显示器的第一轴的更多关键帧。 基于关键帧或关键帧的内容，确定沿着第二轴显示一个或多个关键帧的位置。

公开(公告)号：US20090116811A1

公开(公告)日：2009-05-07

申请号：US12252023

申请日：2008-10-15

申请人： Mayank Kukreja ,  Srinivasan H. Sengamedu

发明人： Mayank Kukreja ,  Srinivasan H. Sengamedu

IPC分类号： H04N5/93 ,  G06K9/34 ,  G06K9/46

CPC分类号： G11B27/34 ,  G06F16/745 ,  G06F16/78 ,  G06F16/784 ,  G06F16/785 ,  G06F16/7857 ,  G06F16/786 ,  G06K9/00711 ,  G11B27/28

摘要： Keyframes of video are arranged on a display based on characteristics on the keyframes, such as content similarity and temporal relation as compared to each other, where input is received comprising one or more keyframes from video data and it is determined where to display the one or more keyframes along a first axis of the display based on a time associated with the keyframe or keyframes. It is determined where to display the one or more keyframes along a second axis based on the content of the keyframe or keyframes.

摘要翻译： 视频的关键帧基于关键帧上的特征被布置在显示器上，例如彼此相比的内容相似性和时间关系，其中接收包括来自视频数据的一个或多个关键帧的输入，并且确定在哪里显示一个或 基于与关键帧或关键帧相关联的时间沿着显示器的第一轴的更多关键帧。 基于关键帧或关键帧的内容，确定沿着第二轴显示一个或多个关键帧的位置。

公开(公告)号：US20130293776A1

公开(公告)日：2013-11-07

申请号：US13935183

申请日：2013-07-03

申请人： The Trustees of Columbia University

发明人： Chang Shih-Fu ,  Dongqing Zhang

IPC分类号： H04N7/088

CPC分类号： H04N7/0882 ,  G06F16/739 ,  G06F16/7844 ,  G06F16/7857 ,  G06F16/786 ,  G06K9/3266 ,  G06K2209/01 ,  G06T7/75 ,  G11B27/031 ,  G11B27/034 ,  G11B27/28 ,  H04N5/93 ,  H04N7/025 ,  H04N21/434 ,  H04N21/4884

摘要： Caption boxes which are embedded in video content can be located and the text within the caption boxes decoded. Real time processing is enhanced by locating caption box regions in the compressed video domain and performing pixel based processing operations within the region of the video frame in which a caption box is located. The captions boxes are further refined by identifying word regions within the caption boxes and then applying character and word recognition processing to the identified word regions. Domain based models are used to improve text recognition results. The extracted caption box text can be used to detect events of interest in the video content and a semantic model applied to extract a segment of video of the event of interest.

摘要翻译： 嵌入在视频内容中的字幕框可以被定位，并且标题框内的文本被解码。 通过将标题框区域定位在压缩视频域中并且在字幕框所在的视频帧的区域内执行基于像素的处理操作来增强实时处理。 通过识别字幕框内的字区域，然后对识别的字区域应用字符和字识别处理，进一步细化字幕框。 基于域的模型用于改进文本识别结果。 提取的字幕框文本可用于检测视频内容中感兴趣的事件和应用于提取感兴趣事件的视频片段的语义模型。

公开(公告)号：US20230297617A1

公开(公告)日：2023-09-21

申请号：US18136538

申请日：2023-04-19

申请人： Tencent Technology (Shenzhen) Company Limited

发明人： Hui Guo

IPC分类号： G06F16/783 ,  G06F16/732 ,  G06V10/44 ,  G06V10/54 ,  G06V10/74

CPC分类号： G06F16/7857 ,  G06F16/7328 ,  G06V10/44 ,  G06V10/54 ,  G06V10/761

摘要： This application provides a video retrieval method performed by a computer device. The method includes: performing feature extraction on an image feature of a query video to obtain a first quantization feature, obtaining a second candidate video with a high category similarity to the query video based on the first quantization feature, and finally taking a second candidate video with a high content similarity to the query video as a target video. The quantization control parameters are adjusted according to the texture feature loss value corresponding to each training sample to cause the target quantization processing sub-model to learn the ranking ability of the target texture feature sub-model, to ensure that the ranking effect of two sub-models tend to be consistent, and an end-to-end model architecture enables the target quantization processing sub-model to obtain the corresponding quantization feature based on the image feature.

公开(公告)号：US20080162561A1

公开(公告)日：2008-07-03

申请号：US11619342

申请日：2007-01-03

申请人： Milind R. Naphade ,  John R. Smith

发明人： Milind R. Naphade ,  John R. Smith

IPC分类号： G06F17/30

CPC分类号： G06F16/785 ,  G06F16/745 ,  G06F16/7854 ,  G06F16/7857 ,  G06F16/786

摘要： An embodiment of the present invention relates to the combining of multiple semantic analyses of audio-visual data in order to resolve a higher fidelity description of the semantic content and more specifically to a method for applying semantic concept detection over multiple related audio-video sources, scoring the sources on the basis of presence or absence of specific semantics and aggregating the scores using combination functions to achieve a semantic super-resolution.

摘要翻译： 本发明的一个实施例涉及对视听数据的多个语义分析的组合，以便解析语义内容的较高保真度描述，更具体地说涉及一种在多个相关音频视频源上应用语义概念检测的方法， 在存在或不存在特定语义的基础上对源进行评分，并使用组合函数聚合分数，以实现语义超分辨率。

公开(公告)号：US20060109908A1

公开(公告)日：2006-05-25

申请号：US11328054

申请日：2006-01-10

申请人： Noboru Yamaguchi ,  Toshiaki Watanabe ,  Takashi Ida ,  Yoko Sambonsugi ,  Osamu Hori ,  Toshimitsu Kaneko

发明人： Noboru Yamaguchi ,  Toshiaki Watanabe ,  Takashi Ida ,  Yoko Sambonsugi ,  Osamu Hori ,  Toshimitsu Kaneko

IPC分类号： H04N11/02 ,  H04N7/12 ,  H04N11/04 ,  H04B1/66

CPC分类号： G06F16/7854 ,  G06F16/7857

摘要： An apparatus for retrieving a video picture includes a decoder section for decoding a coded bit stream of video picture data representing an arbitrary shape object and including shape information and texture information, a retrieval condition input section for inputting a retrieval condition for retrieval of a desired picture, a retrieval section for retrieving a picture meeting the retrieval condition by using shape information of the object decoded by the decoder section, and a display section for outputting the retrieved result obtained by the retrieval section.

公开(公告)号：US20060064716A1

公开(公告)日：2006-03-23

申请号：US11221397

申请日：2005-09-07

申请人： Sanghoon Sull ,  Hyeokman Kim ,  Yeon-Seok Seong ,  Michael Rostoker ,  Jung Kim

发明人： Sanghoon Sull ,  Hyeokman Kim ,  Yeon-Seok Seong ,  Michael Rostoker ,  Jung Kim

IPC分类号： H04N5/445 ,  H04N5/44 ,  G06F17/00 ,  G06F17/21

CPC分类号： G11B27/34 ,  G06F16/739 ,  G06F16/743 ,  G06F16/784 ,  G06F16/785 ,  G06F16/7857 ,  G11B27/105 ,  G11B27/28 ,  G11B2220/20 ,  G11B2220/41 ,  H04N21/4884

摘要： Techniques for poster-thumbnail and/or animated thumbnail development and/or usage to effectively navigate for potential selection between a plurality of images or programs/video files or video segments. The poster and animated thumbnail images are presented in a GUI on adapted apparatus to provide an efficient system for navigating, browsing and/or selecting images or programs or video segments to be viewed by a user. The poster and animated thumbnails may be automatically produced without human-necessary editing and may also have one or more various associated data (such as text overlay, image overlay, cropping, text or image deletion or replacement, and/or associated audio).

摘要翻译： 用于海报缩略图和/或动画缩略图开发和/或使用的技术以有效地导航用于多个图像或节目/视频文件或视频片段之间的潜在选择。 海报和动画缩略图在适配设备上的GUI中呈现，以提供用于导航，浏览和/或选择用户要观看的图像或节目或视频片段的有效系统。 海报和动画缩略图可以在没有人为必要的编辑的情况下自动产生，并且还可以具有一个或多个各种相关联的数据（例如文本覆盖，图像叠加，裁剪，文本或图像删除或替换，和/或相关联的音频）。

公开(公告)号：US20040255249A1

公开(公告)日：2004-12-16

申请号：US10494739

申请日：2004-05-08

发明人： Shih-Fu Chang ,  Dongqing Zhang

IPC分类号： H04N005/14 ,  H04N011/00

CPC分类号： H04N7/0882 ,  G06F16/739 ,  G06F16/7844 ,  G06F16/7857 ,  G06F16/786 ,  G06K9/3266 ,  G06K2209/01 ,  G06T7/75 ,  G11B27/031 ,  G11B27/034 ,  G11B27/28 ,  H04N5/93 ,  H04N7/025 ,  H04N21/434 ,  H04N21/4884

摘要： Caption boxes which are embedded in video content can be located and the text within the caption boxes decoded. Real time processing is enhanced by locating caption box regions in the compressed video domain (210) and performing pixel based processing operations within the region of the video frame in which a caption box is located. The captions boxes are further refined by identifying word regions (240) within the caption boxes and then applying character and word recognition processing (250) to the identified word regions. Domain based models are used to improve text recognition results. The extracted caption box text can be used to detect events of interest in the video content and a semantic model applied to extract a segment of video of the event of interest.

摘要翻译： 嵌入在视频内容中的字幕框可以被定位，并且标题框内的文本被解码。 通过将标题框区域定位在压缩视频域（210）中并且在字幕框所在的视频帧的区域内执行基于像素的处理操作来增强实时处理。 通过识别字幕框内的字区域（240），然后将字符和字识别处理（250）应用到所识别的字区域，来进一步改进字幕框。 基于域的模型用于改进文本识别结果。 提取的字幕框文本可用于检测视频内容中感兴趣的事件和应用于提取感兴趣事件的视频片段的语义模型。