公开(公告)号：EP1579351A4

公开(公告)日：2008-07-02

申请号：EP03813399

申请日：2003-12-12

Applicant: IBM

Inventor： SMITH JOHN ,  NAPHADE MILIND ,  NATSEV APOSTAL

IPC: G06F17/30

CPC classification number: G06F17/3079 ,  G06F17/30247 ,  G06F17/30265 ,  G06F17/30858 ,  G06K9/00624 ,  Y10S707/99931 ,  Y10S707/99943 ,  Y10S707/99945 ,  Y10S707/99948

Abstract: A method and apparatus for extracting a model vector representation from multimedia documents (104). A model vector provides a multidimensional representation of the confidence with which multimedia documents belong to a set of categories or with which a set of semantic concepts relate to the documents. A model vector can be associated with multimedia documents to provide an index of its content or categorization and can be used for comparing, searching, classifying, or clustering multimedia documents. A model vector can be used for purposes of information discovery, personalizing multimedia content, and querying a multimedia information repository (103).

公开(公告)号：WO0228006A3

公开(公告)日：2002-07-18

申请号：PCT/GB0104167

申请日：2001-09-19

Applicant: IBM ,  IBM UK

Inventor： CHANG YUAN-CHI ,  LI CHUNG-SHENG ,  HAN RICHARD YEH-WHEI ,  SMITH JOHN

IPC: G06F13/00 ,  G06F21/24 ,  H04L29/06 ,  H04N7/167 ,  H04N7/26 ,  H04N21/222 ,  H04N21/2343 ,  H04N21/2347 ,  H04N21/235 ,  H04N21/84

CPC classification number: H04L63/0442 ,  H04L29/06027 ,  H04L63/12 ,  H04L65/103 ,  H04L65/104 ,  H04N7/1675 ,  H04N19/40 ,  H04N21/222 ,  H04N21/234309 ,  H04N21/2347 ,  H04N21/2351 ,  H04N21/84

Abstract: A method and system enables network intermediaries such as proxies to transcode multimedia data without violating end-to-end security guarantees. In one embodiment, a content provider decomposes a data stream into a plurality of components, each of which are independently encrypted. In a preferred embodiment, an intermediary or proxy performs transcoding of the components by prioritizing and dropping predetermined components, in accordance with unencrypted clear-text metadata associated with each component, without the need for decrypting the components. Clear-text metadata preferably provides a semantic understanding of the absolute or relative importance/priority of the components with respect to each other, thereby facilitating the transcoding process. The destination/client device can subsequently decrypt the components transmitted by the transcoding proxy and reassemble them into a representation of the original data content.

Abstract translation: 一种方法和系统使得诸如代理的网络中介能够在不违反端对端安全保证的情况下对多媒体数据进行代码转换。 在一个实施例中，内容提供者将数据流分解成多个组件，每个组件都是独立加密的。 在优选实施例中，中介或代理根据与每个组件相关联的未加密明文元数据通过优先化和丢弃预定组件来执行组件的代码转换，而不需要解密组件。 明文元数据优选地提供组件相对于彼此的绝对或相对重要性/优先级的语义理解，从而促进代码转换过程。 目的地/客户端设备可以随后解密由代码转换代理传输的组件，并将它们重新组装成原始数据内容的表示。

公开(公告)号：WO2004055639A3

公开(公告)日：2004-11-04

申请号：PCT/US0339548

申请日：2003-12-12

Applicant: IBM

Inventor： SMITH JOHN ,  NAPHADE MILIND ,  NATSEV APOSTAL

IPC: G06F17/30

CPC classification number: G06F17/3079 ,  G06F17/30247 ,  G06F17/30265 ,  G06F17/30858 ,  G06K9/00624 ,  Y10S707/99931 ,  Y10S707/99943 ,  Y10S707/99945 ,  Y10S707/99948

Abstract: A method and apparatus for extracting a model vector representation from multimedia documents (104). A model vector provides a multidimensional representation of the confidence with which multimedia documents belong to a set of categories or with which a set of semantic concepts relate to the documents. A model vector can be associated with multimedia documents to provide an index of its content or categorization and can be used for comparing, searching, classifying, or clustering multimedia documents. A model vector can be used for purposes of information discovery, personalizing multimedia content, and querying a multimedia information repository (103).

Abstract translation: 一种用于从多媒体文档（104）提取模型向量表示的方法和装置。 模型向量提供多维文档属于一组类别的置信度的多维表示，或者一组语义概念与文档相关。 模型向量可以与多媒体文档相关联，以提供其内容或分类的索引，并可用于比较，搜索，分类或聚类多媒体文档。 模型向量可以用于信息发现，个性化多媒体内容和查询多媒体信息库（103）的目的。

公开(公告)号：AT291318T

公开(公告)日：2005-04-15

申请号：AT01969934

申请日：2001-09-19

Applicant: IBM

Inventor： CHANG YUAN-CHI ,  LI CHUNG-SHENG ,  HAN RICHARD YEH-WHEI ,  SMITH JOHN

IPC: G06F13/00 ,  G06F21/24 ,  H04L29/06 ,  H04N7/167 ,  H04N7/26 ,  H04N21/222 ,  H04N21/2343 ,  H04N21/2347 ,  H04N21/235 ,  H04N21/84

Abstract: A method and system enables network intermediaries such as proxies to transcode multimedia data without violating end-to-end security guarantees. In one embodiment, a content provider decomposes a data stream into a plurality of components, each of which are independently encrypted. In a preferred embodiment, an intermediary or proxy performs transcoding of the components by prioritizing and dropping predetermined components, in accordance with unencrypted clear-text metadata associated with each component, without the need for decrypting the components. Clear-text metadata preferably provides a semantic understanding of the absolute or relative importance/priority of the components with respect to each other, thereby facilitating the transcoding process. The destination/client device can subsequently decrypt the components transmitted by the transcoding proxy and reassemble them into a representation of the original data content.

公开(公告)号：AU2003300881A8

公开(公告)日：2004-07-09

申请号：AU2003300881

申请日：2003-12-12

Applicant: IBM

Inventor： NATSEV APOSTAL ,  NAPHADE MILIND ,  SMITH JOHN

IPC: G06F17/30

Abstract: A method and apparatus for extracting a model vector representation from multimedia documents. A model vector provides a multidimensional representation of the confidence with which multimedia documents belong to a set of categories or with which a set of semantic concepts relate to the documents. A model vector can be associated with multimedia documents to provide an index of its content or categorization and can be used for comparing, searching, classifying, or clustering multimedia documents. A model vector can be used for purposes of information discovery, personalizing multimedia content, and querying a multimedia information repository.

公开(公告)号：AU9005901A

公开(公告)日：2002-04-08

申请号：AU9005901

申请日：2001-09-19

Applicant: IBM

Inventor： LI CHUNG-SHENG ,  HAN RICHARD YEH-WHEI ,  SMITH JOHN

IPC: G06F13/00 ,  G06F21/24 ,  H04L29/06 ,  H04N7/167 ,  H04N7/26 ,  H04N21/222 ,  H04N21/2343 ,  H04N21/2347 ,  H04N21/235 ,  H04N21/84 ,  H04L9/00

Abstract: A method and system enables network intermediaries such as proxies to transcode multimedia data without violating end-to-end security guarantees. In one embodiment, a content provider decomposes a data stream into a plurality of components, each of which are independently encrypted. In a preferred embodiment, an intermediary or proxy performs transcoding of the components by prioritizing and dropping predetermined components, in accordance with unencrypted clear-text metadata associated with each component, without the need for decrypting the components. Clear-text metadata preferably provides a semantic understanding of the absolute or relative importance/priority of the components with respect to each other, thereby facilitating the transcoding process. The destination/client device can subsequently decrypt the components transmitted by the transcoding proxy and reassemble them into a representation of the original data content.

公开(公告)号：DE60209473D1

公开(公告)日：2006-04-27

申请号：DE60209473

申请日：2002-05-03

Applicant: IBM

Inventor： KUIKEN PAUL ,  SMITH JOHN

IPC: G06F3/048 ,  G06F9/44 ,  G06F3/00 ,  G06F3/14 ,  G06F9/40 ,  G06F13/00

Abstract: A message sequencing mechanism adds text in a display area based on relational importance. Each message is coded with an identification. The message is added to the display area based on previous identifications. The messages are categorized and sorted first by identification and then by chronology. Messages may be displayed in ascending or descending order with respect to both the identification and the time order of the messages. If the identification dictates that the message is to be added between two existing entries, a space for the new message is made between related entries in the display area, and the new message is added. Messages may also include a sub identification for more defined relationships. Furthermore, a message may be text or other media, such as sound, image, or video. The field may also be automatically scrolled to display the most recent message in association with related messages.

公开(公告)号：AU2003300881A1

公开(公告)日：2004-07-09

申请号：AU2003300881

申请日：2003-12-12

Applicant: IBM

Inventor： NAPHADE MILIND ,  NATSEV APOSTAL ,  SMITH JOHN

IPC: G06F17/30

Abstract: A method and apparatus for extracting a model vector representation from multimedia documents. A model vector provides a multidimensional representation of the confidence with which multimedia documents belong to a set of categories or with which a set of semantic concepts relate to the documents. A model vector can be associated with multimedia documents to provide an index of its content or categorization and can be used for comparing, searching, classifying, or clustering multimedia documents. A model vector can be used for purposes of information discovery, personalizing multimedia content, and querying a multimedia information repository.

公开(公告)号：DE60209473T2

公开(公告)日：2006-09-21

申请号：DE60209473

申请日：2002-05-03

Applicant: IBM

Inventor： KUIKEN PAUL ,  SMITH JOHN

IPC: G06F3/048 ,  G06F9/44 ,  G06F3/00 ,  G06F3/14 ,  G06F9/40 ,  G06F13/00

Abstract: A message sequencing mechanism adds text in a display area based on relational importance. Each message is coded with an identification. The message is added to the display area based on previous identifications. The messages are categorized and sorted first by identification and then by chronology. Messages may be displayed in ascending or descending order with respect to both the identification and the time order of the messages. If the identification dictates that the message is to be added between two existing entries, a space for the new message is made between related entries in the display area, and the new message is added. Messages may also include a sub identification for more defined relationships. Furthermore, a message may be text or other media, such as sound, image, or video. The field may also be automatically scrolled to display the most recent message in association with related messages.

公开(公告)号：DE60109467T2

公开(公告)日：2006-04-13

申请号：DE60109467

申请日：2001-09-19

Applicant: IBM

Inventor： CHANG YUAN-CHI ,  LI CHUNG-SHENG C O IBM UK LTD ,  HAN RICHARD ,  SMITH JOHN

IPC: G06F13/00 ,  H04L29/06 ,  G06F21/24 ,  H04N7/167 ,  H04N7/26 ,  H04N21/222 ,  H04N21/2343 ,  H04N21/2347 ,  H04N21/235 ,  H04N21/84

Abstract: A method and system enables network intermediaries such as proxies to transcode multimedia data without violating end-to-end security guarantees. In one embodiment, a content provider decomposes a data stream into a plurality of components, each of which are independently encrypted. In a preferred embodiment, an intermediary or proxy performs transcoding of the components by prioritizing and dropping predetermined components, in accordance with unencrypted clear-text metadata associated with each component, without the need for decrypting the components. Clear-text metadata preferably provides a semantic understanding of the absolute or relative importance/priority of the components with respect to each other, thereby facilitating the transcoding process. The destination/client device can subsequently decrypt the components transmitted by the transcoding proxy and reassemble them into a representation of the original data content.