公开(公告)号：US20190215580A1

公开(公告)日：2019-07-11

申请号：US15865173

申请日：2018-01-08

Applicant: NETFLIX, INC.

Inventor： Murthy PARTHASARATHI ,  Ghaida ZAHRAN ,  Devang SHAH ,  Andrew SWAN ,  Thomas E. MACK ,  Kalyan NISTALA

IPC: H04N21/854 ,  G06F17/24 ,  H04N21/488

CPC classification number: H04N21/854 ,  G06F17/248 ,  G06F17/2836 ,  H04N21/4884

Abstract: In various embodiments, a subtitle conformance application causes modifications to a subtitle list based on changes associated with an audiovisual program. In operation, the subtitle conformance application performs comparison operation(s) between versions of a subtitle template to identify changes to subtitles associated with the audiovisual program. The subtitle conformance application then determines a mapping between a first change included in the changes and a subtitle list associated with the audiovisual program. Finally, the subtitle conformance application causes the subtitle list to be modified based on the first change and the mapping. Advantageously, the subtitle conformance application enables productive development of subtitles to begin before the audiovisual program is finalized.

公开(公告)号：US20190215421A1

公开(公告)日：2019-07-11

申请号：US15863772

申请日：2018-01-05

Applicant: NETFLIX, INC.

Inventor： Murthy PARTHASARATHI ,  Andrew SWAN ,  Yadong WANG ,  Thomas E. MACK

IPC: H04N5/14 ,  H04N21/44 ,  H04N21/488 ,  H04N21/43 ,  H04N9/89

CPC classification number: H04N5/147 ,  H04N5/44513 ,  H04N9/8233 ,  H04N9/89 ,  H04N21/23418 ,  H04N21/242 ,  H04N21/4302 ,  H04N21/44008 ,  H04N21/4884 ,  H04N2005/44526

Abstract: In various embodiments, a subtitle timing application detects timing errors between subtitles and shot changes. In operation, the subtitle timing application determines that a temporal edge associated with a subtitle does not satisfy a timing guideline based on a shot change. The shot change occurs within a sequence of frames of an audiovisual program. The subtitle timing application then determines a new temporal edge that satisfies the timing guideline relative to the shot change. Subsequently, the subtitle timing application causes a modification to a temporal location of the subtitle within the sequence of frames based on the new temporal edge. Advantageously, the modification to the subtitle improves a quality of a viewing experience for a viewer. Notably, by automatically detecting timing errors, the subtitle timing application facilitates proper and efficient re-scheduling of subtitles that are not optimally timed with shot changes.

公开(公告)号：US10332116B2

公开(公告)日：2019-06-25

申请号：US14876613

申请日：2015-10-06

Applicant: NETFLIX, INC.

Inventor： Rudra Peram ,  Jason Chan

IPC: G06Q40/00 ,  G06Q20/40 ,  H04L29/08 ,  G06Q20/32 ,  G06Q30/02

Abstract: Provided herein are systems and methods of monitoring account activity in a streaming media environment. An exemplary system includes a monitoring system, an account creation and management system, and an account payment system. The monitoring system is coupled to the account creation and management system and the account payment system via a network. The processing device of the monitoring system retrieves account information for a first user account. Account information includes user consumption information and user payment information associated with the first user account. The processing device determines a fraudulent account score for the first user account based on at least one of the user consumption information, the user payment information, and account identification information. When the fraudulent account score exceeds an upper threshold, the processing device automatically deletes the first user account from at least one of the account creation and management system and the accounts payment system.

公开(公告)号：US10306270B2

公开(公告)日：2019-05-28

申请号：US15633661

申请日：2017-06-26

Applicant: NETFLIX, Inc.

Inventor： Frederic Turmel ,  Nicholas Newell

IPC: H04N21/234 ,  H04N19/89 ,  H04L29/06 ,  H04L12/26 ,  H04L12/24

Abstract: An endpoint device outputs frames of test media during a testing procedure. Each frame of test media includes a test pattern. A test module coupled to the endpoint device samples the test pattern and transmits sample data to a media test engine. The media test engine decodes a binary number from the test pattern and then converts the binary number to an integer value that is associated with the corresponding frame. The media test engine then analyzes sequences of these integer values to identify playback errors associated with the endpoint device.

公开(公告)号：US10263890B2

公开(公告)日：2019-04-16

申请号：US15237541

申请日：2016-08-15

Applicant: Netflix, Inc.

Inventor： David Bannister

IPC: H04L12/741 ,  H04L12/733 ,  H04L12/715 ,  H04L12/753

Abstract: One embodiment of the present invention sets forth a technique for compressing a forwarding table. The technique includes selecting, from a listing of network prefixes, a plurality of network prefixes that are within a range of a subnet. The technique further includes sorting the plurality of network prefixes to generate one or more subgroups of network prefixes and selecting a first subgroup of network prefixes included in the one or more subgroups of network prefixes. The technique further includes generating a synthetic supernet based on the first subgroup of network prefixes.

公开(公告)号：US20190109788A1

公开(公告)日：2019-04-11

申请号：US16209896

申请日：2018-12-04

Applicant: NETFLIX, INC.

Inventor： David Bannister

IPC: H04L12/741 ,  H04L12/733 ,  H04L12/715 ,  H04L12/753

Abstract: One embodiment of the present invention sets forth a technique for compressing a forwarding table. The technique includes selecting, from a listing of network prefixes, a plurality of network prefixes that are within a range of a subnet. The technique further includes sorting the plurality of network prefixes to generate one or more subgroups of network prefixes and selecting a first subgroup of network prefixes included in the one or more subgroups of network prefixes. The technique further includes generating a synthetic supernet based on the first subgroup of network prefixes.

公开(公告)号：US20190057195A1

公开(公告)日：2019-02-21

申请号：US16167176

申请日：2018-10-22

Applicant: NETFLIX, INC.

Inventor： Mark Watson ,  Anthony Neal Park ,  Mitch Zollinger

IPC: G06F21/10 ,  H04N21/8355 ,  H04N21/482 ,  H04N21/466 ,  H04N21/4405 ,  H04N21/84 ,  H04N21/6334

Abstract: Techniques are disclosed for improving user experience of multimedia streaming over computer networks. More specifically, techniques presented herein reduce (or eliminate) latency in playback start time for streaming digital media content resulting from digital rights management (DRM) authorizations. A streaming media client (e.g., a browser, set-top box, mobile telephone or tablet “app”) may request a “fast-expiring” license for titles the streaming media client predicts a user is likely to begin streaming. A fast-expiring license is a DRM license (and associated decryption key) which is valid for only a very limited time after being used for playback. During the validity period of such a license, the client device requests a “normal” or “regular” license to continue accessing the title after the fast-expiring license expires.

公开(公告)号：US20190028529A1

公开(公告)日：2019-01-24

申请号：US16053754

申请日：2018-08-02

Applicant: NETFLIX, INC.

Inventor： Ioannis KATSAVOUNIDIS

IPC: H04L29/06 ,  H04N21/2662 ,  H04L12/26

Abstract: A shot analyzer varies the resolution when generating encoded video sequences for streaming. The shot analyzer generates a first encoded video sequence based on a first resolution and a source video sequence that is associated with a video title. The shot analyzer then determines a first encoded shot sequence from multiple encoded shot sequences included in the first encoded video sequence based on quality metric(s). The first encoded shot sequence is associated with a first shot sequence included in the source video sequence. Subsequently, the shot analyzer generates a second encoded shot sequence based on a second resolution and the first shot sequence. The shot analyzer generates a second encoded video sequence based on the first encoded video sequence and the second encoded shot sequence. At least a first portion of the second encoded video sequence is subsequently streamed to an endpoint device during playback of the video title.

公开(公告)号：US10182090B2

公开(公告)日：2019-01-15

申请号：US14846723

申请日：2015-09-04

Applicant: NETFLIX, Inc

Inventor： David Fullagar ,  Kenneth W. Florance ,  Ian Van Hoven

IPC: G06F15/167 ,  H04L29/06 ,  H04N21/218 ,  H04N21/222 ,  H04N21/237 ,  H04N21/25 ,  H04N21/258 ,  H04N21/61 ,  H04L29/08 ,  G06F17/30 ,  G06Q10/00

Abstract: One embodiment of the present invention sets forth a method for updating content stored in a cache residing at an internet service provider (ISP) location that includes receiving popularity data associated with a first plurality of content assets, where the popularity data indicate the popularity of each content asset in the first plurality of content assets across a user base that spans multiple geographic regions, generating a manifest that includes a second plurality of content assets based on the popularity data and a geographic location associated with the cache, where each content asset included in the manifest is determined to be popular among users proximate to the geographic location or users with preferences similar to users proximate to the geographic location, and transmitting the manifest to the cache, where the cache is configured to update one or more content assets stored in the cache based on the manifest.

公开(公告)号：US10180968B2

公开(公告)日：2019-01-15

申请号：US15044020

申请日：2016-02-15

Applicant: NETFLIX, INC.

Inventor： Harald Steck

IPC: G06F17/30 ,  G06N99/00

Abstract: In one embodiment of the present invention, a training engine teaches a matrix factorization model to rank items for users based on implicit feedback data and a rank loss function. In operation, the training engine approximates a distribution of scores to corresponding ranks as an approximately Gaussian distribution. Based on this distribution, the training engine selects an activation function that smoothly maps between scores and ranks. To train the matrix factorization model, the training engine directly optimizes the rank loss function based on the activation function and implicit feedback data. By contrast, conventional training engines that optimize approximations of the rank loss function are typically less efficient and produce less accurate ranking models.