-
公开(公告)号:US10528821B2
公开(公告)日:2020-01-07
申请号:US15689193
申请日:2017-08-29
Applicant: Amazon Technologies, Inc.
Inventor: Adam Carlson , Douglas Ryan Gray , Ashutosh Vishwas Kulkarni , Colin Jon Taylor
Abstract: A video segmentation system can be utilized to automate segmentation of digital video content. Features corresponding to visual, audio, and/or textual content of the video can be extracted from frames of the video. The extracted features of adjacent frames are compared according to a similarity measure to determine boundaries of a first set of shots or video segments distinguished by abrupt transitions. The first set of shots is analyzed according to certain heuristics to recognize a second set of shots distinguished by gradual transitions. Key frames can be extracted from the first and second set of shots, and the key frames can be used by the video segmentation system to group the first and second set of shots by scene. Additional processing can be performed to associate metadata, such as names of actors or titles of songs, with the detected scenes.
-
公开(公告)号:US09436876B1
公开(公告)日:2016-09-06
申请号:US14577277
申请日:2014-12-19
Applicant: Amazon Technologies, Inc.
Inventor: Adam Carlson , Douglas Ryan Gray , Ashutosh Vishwas Kulkarni , Colin Jon Taylor
CPC classification number: G06K9/00765 , G06K9/00147 , G06K9/00744 , G06K9/46 , G06K9/469 , G06K9/6212 , G06K9/6224 , G06T2207/10016
Abstract: A video segmentation system can be utilized to automate segmentation of digital video content. Features corresponding to visual, audio, and/or textual content of the video can be extracted from frames of the video. The extracted features of adjacent frames are compared according to a similarity measure to determine boundaries of a first set of shots or video segments distinguished by abrupt transitions. The first set of shots is analyzed according to certain heuristics to recognize a second set of shots distinguished by gradual transitions. Key frames can be extracted from the first and second set of shots, and the key frames can be used by the video segmentation system to group the first and second set of shots by scene. Additional processing can be performed to associate metadata, such as names of actors or titles of songs, with the detected scenes.
Abstract translation: 可以利用视频分割系统来自动分割数字视频内容。 可以从视频的帧中提取与视频的视觉,音频和/或文本内容相对应的特征。 根据相似性度量来比较相邻帧的提取特征,以确定通过突变转换区分的第一组镜头或视频片段的边界。 根据某些启发式分析第一组镜头,以识别通过逐渐转换区分的第二组镜头。 可以从第一和第二组拍摄中提取关键帧,并且关键帧可以被视频分割系统用于按照场景对第一组和第二组拍摄进行分组。 可以执行附加处理以将元数据(例如演员的名称或歌曲的名称)与检测到的场景相关联。
-
公开(公告)号:US11941680B1
公开(公告)日:2024-03-26
申请号:US16551537
申请日:2019-08-26
Applicant: Amazon Technologies, Inc.
Inventor: Stephen Brent Ivie , Ashutosh Vishwas Kulkarni , Saurabh Nangia , Adam Landry Bordelon , Aaron James Dykstra , David Michael Hurley , Adam James Finkelstein , Scott James McKee
IPC: G06Q30/0601 , G06Q30/06 , H04L9/40
CPC classification number: G06Q30/0631 , G06Q30/06 , G06Q30/0641 , H04L63/08
Abstract: Social network postings, including text, images or other media, may provide valuable information regarding a user of the social network with which the postings may be associated. With the authorization of the user, and upon authentication by the social network, an online marketplace may access the social network postings and extract data therefrom, and market one or more recommended items to the user based on the extracted data, which may include color pallets or texture pallets derived from photographs included in the postings.
-
公开(公告)号:US10395297B1
公开(公告)日:2019-08-27
申请号:US13675333
申请日:2012-11-13
Applicant: Amazon Technologies, Inc.
Inventor: Stephen Brent Ivie , Ashutosh Vishwas Kulkarni , Saurabh Nangia , Adam Landry Bordelon , Aaron James Dykstra , David Michael Hurley , Adam James Finkelstein , Scott James McKee
Abstract: Social network postings, including text, images or other media, may provide valuable information regarding a user of the social network with which the postings may be associated. With the authorization of the user, and upon authentication by the social network, an online marketplace may access the social network postings and extract data therefrom, and market one or more recommended items to the user based on the extracted data, which may include color pallets or texture pallets derived from photographs included in the postings.
-
公开(公告)号:US20170083770A1
公开(公告)日:2017-03-23
申请号:US15255978
申请日:2016-09-02
Applicant: Amazon Technologies, Inc.
Inventor: Adam Carlson , Douglas Ryan Gray , Ashutosh Vishwas Kulkarni , Colin Jon Taylor
CPC classification number: G06K9/00765 , G06K9/00147 , G06K9/00744 , G06K9/46 , G06K9/469 , G06K9/6212 , G06K9/6224 , G06T2207/10016
Abstract: A video segmentation system can be utilized to automate segmentation of digital video content. Features corresponding to visual, audio, and/or textual content of the video can be extracted from frames of the video. The extracted features of adjacent frames are compared according to a similarity measure to determine boundaries of a first set of shots or video segments distinguished by abrupt transitions. The first set of shots is analyzed according to certain heuristics to recognize a second set of shots distinguished by gradual transitions. Key frames can be extracted from the first and second set of shots, and the key frames can be used by the video segmentation system to group the first and second set of shots by scene. Additional processing can be performed to associate metadata, such as names of actors or titles of songs, with the detected scenes.
-
公开(公告)号:US20180082127A1
公开(公告)日:2018-03-22
申请号:US15689193
申请日:2017-08-29
Applicant: Amazon Technologies, Inc.
Inventor: Adam Carlson , Douglas Ryan Gray , Ashutosh Vishwas Kulkarni , Colin Jon Taylor
CPC classification number: G06K9/00765 , G06K9/00147 , G06K9/00744 , G06K9/46 , G06K9/469 , G06K9/6212 , G06K9/6224 , G06T2207/10016
Abstract: A video segmentation system can be utilized to automate segmentation of digital video content. Features corresponding to visual, audio, and/or textual content of the video can be extracted from frames of the video. The extracted features of adjacent frames are compared according to a similarity measure to determine boundaries of a first set of shots or video segments distinguished by abrupt transitions. The first set of shots is analyzed according to certain heuristics to recognize a second set of shots distinguished by gradual transitions. Key frames can be extracted from the first and second set of shots, and the key frames can be used by the video segmentation system to group the first and second set of shots by scene. Additional processing can be performed to associate metadata, such as names of actors or titles of songs, with the detected scenes.
-
公开(公告)号:US09564177B1
公开(公告)日:2017-02-07
申请号:US14667645
申请日:2015-03-24
Applicant: Amazon Technologies, Inc.
Inventor: Douglas Ryan Gray , Adam Carlson , Ashutosh Vishwas Kulkarni , Anna Makris , Colin Jon Taylor
CPC classification number: G11B27/3081 , G11B27/007 , G11B27/105 , H04N5/783 , H04N5/85 , H04N9/8205
Abstract: Automatic replay or skip ahead functionality can be configured to intelligently navigate to a portion of a video a user desires to view. The context at which a user selects intelligent navigation can be analyzed to determine where to initiate automatic replay or skip ahead. The context for intelligent navigation can be based on scene or shot segmentation data, closed captioning, aggregate video navigation data from a community of users of shared demographic traits and/or interest, and/or other metadata. In the case of automatic replay, playback of a portion of a video can include enhancements for that portion, such as providing closed captioning, display at a decreased frame rate (“slow motion”), zooming in/out on a portion of the frames of a video segment, among other enhancements.
Abstract translation: 自动重播或跳过功能可以配置为智能地导航到用户希望查看的视频的一部分。 可以分析用户选择智能导航的上下文以确定在何处启动自动重放或跳过。 智能导航的背景可以基于场景或拍摄分割数据,隐藏字幕,来自共享人口特征和/或兴趣的用户社区的聚合视频导航数据和/或其他元数据。 在自动重放的情况下,视频的一部分的回放可以包括对该部分的增强,例如提供隐藏字幕,以降低的帧速率(“慢动作”)显示,在帧的一部分上放大/缩小 的视频片段,以及其他增强功能。
-
公开(公告)号:US09558784B1
公开(公告)日:2017-01-31
申请号:US14667652
申请日:2015-03-24
Applicant: Amazon Technologies, Inc.
Inventor: Douglas Ryan Gray , Adam Carlson , Ashutosh Vishwas Kulkarni , Anna Makris , Colin Jon Taylor
CPC classification number: G11B27/005 , H04N5/783 , H04N9/8205
Abstract: Automatic replay or skip ahead functionality can be configured to intelligently navigate to a portion of a video a user desires to view. The context at which a user selects intelligent navigation can be analyzed to determine where to initiate automatic replay or skip ahead. The context for intelligent navigation can be based on scene or shot segmentation data, closed captioning, aggregate video navigation data from a community of users of shared demographic traits and/or interest, and/or other metadata. In the case of automatic replay, playback of a portion of a video can include enhancements for that portion, such as providing closed captioning, display at a decreased frame rate (“slow motion”), zooming in/out on a portion of the frames of a video segment, among other enhancements.
-
公开(公告)号:US09805270B2
公开(公告)日:2017-10-31
申请号:US15255978
申请日:2016-09-02
Applicant: Amazon Technologies, Inc.
Inventor: Adam Carlson , Douglas Ryan Gray , Ashutosh Vishwas Kulkarni , Colin Jon Taylor
CPC classification number: G06K9/00765 , G06K9/00147 , G06K9/00744 , G06K9/46 , G06K9/469 , G06K9/6212 , G06K9/6224 , G06T2207/10016
Abstract: A video segmentation system can be utilized to automate segmentation of digital video content. Features corresponding to visual, audio, and/or textual content of the video can be extracted from frames of the video. The extracted features of adjacent frames are compared according to a similarity measure to determine boundaries of a first set of shots or video segments distinguished by abrupt transitions. The first set of shots is analyzed according to certain heuristics to recognize a second set of shots distinguished by gradual transitions. Key frames can be extracted from the first and second set of shots, and the key frames can be used by the video segmentation system to group the first and second set of shots by scene. Additional processing can be performed to associate metadata, such as names of actors or titles of songs, with the detected scenes.
-
-
-
-
-
-
-
-