Apple Inc.

John Y Su

David L Biderman

Roger N Pantos

Hsi-Jung Wu

Xiaosong Zhou

William B May, Jr

Yeping Su

Chris Yoochang Chung

James David Batson

Christopher L Flick

Douglas Scott Price

Christopher L Oklota

Zhenheng Li

Hassan A Shojania

David Alan Pound

Jordan B Schneider

Srinivasan Nimmala

Srirang Ashok Lovlekar

Jianxiong Shi

Murtaza A Shikari

Ajoy K Singh

Jun Xin

Faraz Faheem

Samir Gehani

Courtney Ann Kennedy

Fanyi Duanmu

Karthik Chockalingam

Michel A Rynderman

Jacob A Baron

Eunhee Lim

Soutik Chakraborty

Eduardo Asbun

Shalini Sahoo

Ashkan Heshmati

Suresh Koppisetty

A media streaming method is disclosed in which a network environment of a sink device engaged in media streaming is estimated and at least two network throughput estimates are developed. A first network throughput estimate may be estimated from a measurement of network performance and a second network throughput estimate may be developed from a correlation of the estimated network environment to a machine learning model representing network throughput predictions. A final throughput estimate may be developed from the first and second network throughput estimates; and a representation of media content may be selected for retrieval based on the final throughput estimate. The machine learning model of network throughput may be developed over the course of prior media streaming session(s) that are performed by the sink device in which network throughput performance indicators of the streaming session(s) are stored over a predetermined interval and, upon conclusion of the interval, the model of network throughput is constructed according to a machine learning technique. Both the logging network throughput performance indicators and the building of the model of network throughput may be performed solely by the sink device, which preserves confidentiality of data representing consumer behavior during those media streaming sessions.

On-device machine learning-based network bandwidth prediction to improve adaptive media streaming performance

Techniques are disclosed for coding and delivering multi-view video in which the video is represented as a manifest file identifying a plurality of segments of the video available for download. The multi-view video may be partitioned spatially into a plurality of tiles that, in aggregate, encompass the entire spatial area of the video. The tiles are coded as segments contains coded video representing content contained within its respective tile. Tiles may be given different sizes based on saliency of the content within their respective regions. In this manner, tiles with high levels of interest may have relatively large spatial areas, which can lead to efficient coding in the presence of content motion.

Versatile tile coding for multi-view video streaming

An accessory device may establish a short range link to a companion device, and the accessory device may use the companion device as a proxy to conduct a data session over a cellular network. Based on a latency requirement associated with the data session and/or a signal strength of the short range link, the accessory device may dynamically determine to transition its cellular radio from a powered-off mode to a low power mode.

Reduction in latency for cellular handover in wearable devices

Disclosed are systems, methods, devices and computer-readable media for image fetching for timeline scrubbing of digital media. In some implementations, a method comprises: receiving at a first time prior to receiving a scrub command, a first set of scrub images associated with digital media, the first set of scrub images having a first set of positions on a timeline of the digital media; receiving a first scrub command; receiving at a second time after the first time, a second set of scrub images associated with the digital media, the second set of scrub images having a second set of positions on the timeline that fill time gaps in the first set of positions on the timeline; animating, a timeline overlay including the timeline, a playhead and a scrub image window; and selecting a scrub image from the first or second sets of scrub images for presentation in the scrub image window.

Image fetching for timeline scrubbing of digital media

System and methods for improved playback of a video stream are presented. Video snippets are identified that include a number of consecutive frames for playback. Snippets may be evenly temporally spaced in the video stream or may be content adaptive. Then the first frame of a snippet may be selected as the first frame of a scene or other appropriate stopping point. Scene detection, object detection, motion detection, video metadata, or other information generated during encoding or decoding of the video stream may aid in appropriate snippet selection.

Trick play in digital video streaming

Methods and systems for switching between media streams can use a comparison of an average bandwidth for a new stream to an observed bit rate for an existing stream in order to determine whether to switch to the new stream. A check on the amount of data buffered for the existing stream can also be performed in order to determine whether a system should switch to the new stream.

Switching between media streams

System and methods for storing streamed media data to memory for future playback are provided. A device may stream media data associated with a movie over a network connection. The data streamed to the device will be saved to device memory and will then be playable offline. A manifest associated with the media data will be downloaded to the device and modified to reflect the location of the locally stored media data. During subsequent playback of the media data, the manifest will indicate that at least a portion of the media data exists in local memory. Some portions of the media data not stored locally may be subsequently retrieved from a remote source. The stored data may be updated or added to as necessary.

Movie package file format to persist HLS onto disk

A method of streaming media may include opening a single connection between a client device and a media source; requesting a first media segment of a media stream corresponding to selected tier of a media title over the single connection; determining a download bitrate of the single connection; comparing the download bitrate to a predetermined bitrate of the media title; and in response to determining that the download bitrate is less than the predetermined bitrate, opening a plurality of connections between the client device and media source. The predetermined bitrate may be a predetermined bitrate of a next higher bitrate tier in a group of tiers containing the selected tier when the selected tier is not a highest bitrate tier in the group, and a predetermined bitrate of the selected tier otherwise. The method may further include, in response to determining that an aggregate download bitrate of the plurality of connections is less than a multiple of a download bitrate of an individual one of the connections, closing some of the plurality of connections so that only the individual connection remains.

Adaptive connection switching

Disclosed are systems, methods, devices and computer-readable mediums for image fetching for timeline scrubbing of digital media. In some implementations, a method comprises: receiving at a first time prior to receiving a scrub command, a first set of scrub images associated with digital media, the first set of scrub images having a first set of positions on a timeline of the digital media; receiving a first scrub command; receiving at a second time after the first time, a second set of scrub images associated with the digital media, the second set of scrub images having a second set of positions on the timeline that fill time gaps in the first set of positions on the timeline; animating, a timeline overlay including the timeline, a playhead and a scrub image window; and selecting a scrub image from the first or second sets of scrub images for presentation in the scrub image window.

A media delivery service may include a media source, a proxy cache and one or more client devices. The media source may generate coded segments from an input media stream and develop a playlist identifying network locations of the segments. The media source may broadcast the coded segments and their storage locations to a proxy cache in a broadcast channel. The proxy cache may store successfully received coded segments and develop a mapping table identifying the segments' network locations. The client may request segments from the proxy cache based on the received playlist form the media source. If the proxy cache does not store a requested segment, it may so notify the client and the client may request the segment from the media source.

Media delivery service protocol to support large numbers of client with error failover processes

An embodiment of a method of requesting a plurality of media streams can include, as performed by a computing device, receiving a control input to control a playback function of at least one of the plurality of media streams, determining a download bandwidth between the computing device and a media source on a network, classifying the at least one of the plurality of media streams based on the control input and total download bandwidth, and sending a plurality of requests for the plurality media streams over a single connection to the media source, wherein at least one of an order or a content of the requests is based on the classification. The sending of the media stream requests can utilize at least one of a server-side protocol for multiplexing media streams over a single connection or one or more of HTTP pipelining or HTTP range requests over a single connection.

Multi-stream scheduling and requests

Growing community of inventors

Cupertino, CA, United States of America

John Y Su