Telefonaktiebolaget Lm Ericsson (publ)

Zhuangfei Wu

Clinton Priddle

Per Fröjdh

Thomas Rusert

Rickard Sjöberg

Torbjörn Einarsson

Thorsten Lohmar

Kenneth Andersson

Andrey Norkin

Methods and arrangements in a client node and a server node support a decision on decoding and playout of a certain content in the client node. The methods and arrangements relate to the determining, conveying and obtaining of information related to the characteristics (level) required for decoding and playing out the certain media content at a non-regular playout rate. The method and arrangement in the client node further relate to determining, based on the obtained information, which playout rate(s) is/are supported by the client node in regard of level. The methods and arrangements enable the client node to refrain from attempting to decode and play out of the content at a non-regular rate requiring a level which is not supported by the client node.

Method and arrangement for supporting playout of content

In one aspect, a compressed video bit stream is received and divided into packets that comprise either video data or supplemental information. Each packet is marked with a first subset identifier associated with a corresponding bit stream subset. A first sequence parameter set (SPS) is marked with the same first subset identifier as its associated bit stream subset. The first SPS comprises a second subset identifier indicating a decoding dependency of the bit stream subset associated with the first subset identifier on a bit stream subset associated with the second subset identifier. In another aspect, the packets from the bit stream are received and the first SPS is extracted. The first and second subset identifiers are used as relevant subset identifiers, and for each received packet, the first subset identifier is inspected and the packet is extracted when the first subset identifier matches one of the relevant subset identifiers.

Indicating bit stream subsets

The embodiments relates to motion vector prediction and decoding for multi-view video content from multiple camera views (). At least one candidate motion vector predictor () is provided for a current motion vector () to be encoded. A motion vector predictor () is then determined from the at least one candidate motion vector predictor (). Correspondingly, during decoding a motion vector () is determined for a current pixel block () based on a motion vector predictor () determined for the current pixel block ().

Motion vector processing

Methods and arrangements for video coding are provided. In one exemplary embodiment, a method performed by a video decoder may include decoding an encoded video block based on a single index corresponding to the encoded video block and a predetermined reference picture list to obtain a decoded video block. Further, the index may represent a combination of a reference mode and one or more reference pictures associated with the predetermined reference picture list.

Method and arrangement for video coding

Supplemental processing information defining post-decoding instructions for processing decoded media content is co-organized in a file together with a pseudo-identifier of a media codec employed for encoding the media content. The co-organization trigger abortion of decoding of the media content by a legacy media terminal that does not recognize the pseudo-identifier but enables decoding of the media content by a media terminal that recognizes the pseudo-identifier to thereby enable post-decoding processing of the decoded media content using the supplemental processing information. The co-organization thereby prevents legacy media terminals that cannot use the supplemental processing information from decoding the media content and rendering the media content in a media presentation.

Provision of supplemental processing information

The embodiments use prediction weights in connection with predictive encoding and decoding of a depth map in the form of a frame () with pixels () having pixel values representing distances between a camera () and objects. The prediction weight is calculated based on depth clipping plane parameter values associated with the frame () and with a reference frame () based on which pixels of the frame () are encoded and decoded. Using depth clipping plane parameter values when calculating prediction weights, improves coding efficiency of depth map coding and decoding.

Depth map encoding and decoding

The embodiments relates to motion vector prediction and decoding for multi-view video content from multiple camera views (). At least one candidate motion vector predictor () is provided for a current motion vector () to be encoded. A motion vector predictor is then determined from the at least one candidate motion vector predictor () multiplied by a respective scaling factor determined based on baseline distances between camera views (). Correspondingly, during decoding a motion vector () is determined for a current pixel block () based on a motion vector predictor (determined for the current pixel block ().

Metadata defining decoding and rendering instructions for media content to be co-rendered in a media presentation is divided and distributed as track fragments provided in different media container files. Track fragment adjustment information is included in at least one such track fragment in order to define rendering timing relationships between media content portions defined by the track fragments in a current media container file. The rendering timing relationships enable a correct time alignment of the playback of the media content to be co-rendered to achieve a synchronized media presentation. The track fragment adjustment information is particularly advantageous in connection with tuning in or a random access in a stream of media container files comprising fragmented metadata.

File format for synchronized media

In a method of enabling representation switching during HTTP streaming sessions in a communication system, arranging available representations into groups, providing information identifying the groups and their respective representations, and switching representation based on the provided group identify information.

Method and arrangement for representation switching in HTTP streaming

A technique for encoding and decoding data pertaining to multiple video views is presented. A method embodiment of the encoding technique comprises the steps of providing first picture stream belonging to a first view and performing a first encoding operation to encode a picture of the first picture stream. The first encoding operation is based on a single-view video codec and provides encoder state information relevant for the encoding of another picture of the first picture stream. Also provided is a second picture stream belonging to a second video view. A second encoding operation is performed to encode a picture of the second picture stream based on the single-view video codec, wherein the second encoding operation is based on the encoder state information provided by the first encoding operation. On the basis of the encoded picture of the first picture stream and encoded picture of the second picture stream, dedicated video streams are generated.

Multi-view encoding and decoding technique based on single-view video codecs

Supplemental processing information () defining post-decoding instructions for processing decoded media content is co-organized in a file () together with a pseudo-identifier of a media codec employed for encoding the media content. The co-organization trigger abortion of decoding of the media content by a legacy media terminal () that does not recognize the pseudo-identifier but enables decoding of the media content by a media terminal () that recognizes the pseudo-identifier to thereby enable post-decoding processing of the decoded media content using the supplemental processing information (). The co-organization thereby prevents legacy media terminals () that cannot use the supplemental processing information from decoding the media content and rendering the media content in a media presentation.

A method of indicating bit stream subsets in a video bit stream () is provided. The method comprises receiving the bit stream, dividing the bit stream into video packets (-), wherein each packet comprises either one of video data or supplemental information, and marking each packet with a single subset identifier (stream_id). Each subset identifier is associated with a corresponding bit stream subset (-). Further, a method of extracting video packets from a video bit stream is provided. The method comprises providing relevant subset identifiers, receiving video packets from the bit stream, and, for each received packet, inspecting the subset identifier of the packet. The packet is extracted if the subset identifier matches one of the relevant subset identifiers. This allows condensing properties of a bit stream subset into a single identifier, thereby simplifying the processing of video packets in the network and on the client side. Further, devices corresponding to the aforementioned methods are provided.

Growing community of inventors

Danderyd, Sweden

Zhuangfei Wu