Streaming Video SEI Separation for Bandwidth-Efficient Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video streaming technologies waste bandwidth by transmitting Supplemental Enhancement Information (SEI) messages with each coded video picture, even if the information is common across multiple pictures, leading to redundant data transmission and inefficient use of bandwidth.
Innovation Solution
Decoupling SEI messages from coded video pictures, allowing them to be transmitted separately and only when necessary, enabling efficient transmission of sequence-level SEI messages before the corresponding video data, reducing redundant data transfer and improving bandwidth utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If SEI messages are transmitted with coded video data, then decoding accuracy is improved, but bandwidth efficiency deteriorates due to repeated transmission of common information
Solution Approach 1:
The patent segments the video bitstream into separate components: coded video data packets and SEI message packets. This segmentation allows SEI messages to be transmitted independently from coded video data, enabling selective transmission and reuse of common information across multiple pictures without redundant transmission, thereby improving bandwidth efficiency while maintaining decoding accuracy.
Solution Approach 2:
The patent extracts SEI messages from the coded video data stream and transmits them separately. By taking out SEI messages that contain common information applicable to multiple pictures, the system avoids repeated transmission of this information with each coded video picture, reducing bandwidth consumption while ensuring decoding accuracy is maintained through separate message delivery.
2Loss of energy
If SEI messages are transmitted separately from coded video pictures, then bandwidth efficiency is improved, but data synchronization becomes more complex
Solution Approach 1:
The patent implements preliminary action by transmitting SEI messages before their associated coded video pictures. This advance transmission allows the receiving device to prepare and store SEI messages in advance, reducing the synchronization burden during actual decoding. The timing information in SEI messages enables the receiver to know when to apply specific messages, simplifying the overall synchronization process despite separate transmission.
Solution Approach 2:
The patent uses timing information and picture parameter sets as feedback mechanisms to maintain synchronization between separately transmitted SEI messages and coded video pictures. The receiving device uses this feedback information to correctly associate SEI messages with their corresponding pictures, managing synchronization complexity through structured information exchange rather than complex coordination protocols.
3Productivity
If sequence data sets are decoupled from coded video pictures, then streaming efficiency is improved, but file structure complexity increases
Solution Approach 1:
The patent segments the video file into distinct components: sequence data sets containing parameter sets and SEI messages, and coded video picture data. This segmentation enables independent transmission and processing of sequence data, improving streaming efficiency by allowing receivers to acquire necessary parameters and messages before or during video picture reception, while the modular structure manages file complexity through clear separation of concerns.
Solution Approach 2:
The patent creates universal sequence data sets that can serve multiple coded video pictures through the use of picture parameter sets and timing information. These decoupled sequence data sets function as reusable templates that apply to multiple pictures, improving streaming efficiency by avoiding redundant data transmission while the standardized multi-functional structure manages file complexity through versatile, reusable data elements.
Data Source
AI summary
A device may encapsulate video data such that Supplemental Enhancement Information (SEI) messages are stored separately from a sequence of coded video pictures described by the SEI messages. An example device includes a control unit configured to generate one or more SEI messages separate from the coded video pictures, wherein the SEI messages describe respective ones of the sequence of coded video pictures and include elements common to more than one of the coded video pictures, and an output interface configured to output the SEI messages separately from the sequence of coded video pictures. An example destination device may receive the SEI messages separately from the coded video pictures and render the coded video pictures using the SEI messages.