Video Coding Temporal Reference Parameter Signaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing Hypothetical Reference Decoder (HRD) design relies on external presentation time, making it difficult to derive relative presentation time for pictures in coded video sequences other than the first, leading to suboptimal performance, especially when bitstreams are spliced from different sequences.
Innovation Solution
Signaling a parameter indicative of the temporal difference between the last picture of one coded video sequence and the first picture of a subsequent sequence, either by modifying the semantics of the syntax element for temporal reference or by explicitly signaling the temporal gap between IDR pictures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the HRD relies on external presentation time, then the decoding process can be simplified, but it becomes difficult to derive relative presentation time for pictures in coded video sequences other than the first
Solution Approach 1:
The patent applies preliminary action by embedding temporal reference information (such as picture_distance or temporal gap parameters) into the bitstream during encoding. This allows the decoder to derive presentation times without relying on external sources, as the necessary temporal reference data is prepared and included in advance within the coded video sequence structure.
2Adaptability or versatility
If multiple coded video sequences are combined in a bitstream, then more content can be delivered, but temporal referencing becomes inaccurate without additional parameters
Solution Approach 1:
The patent introduces intermediary parameters (picture_distance, temporal gap, or temporal difference values) that act as mediators between multiple coded video sequences. These parameters bridge the temporal reference information across sequence boundaries, enabling accurate timing relationships to be maintained even when sequences are spliced or combined in the bitstream.
3Device complexity
If the temporal gap between sequences is not signaled, then the bitstream structure remains simple, but the HRD cannot operate optimally when sequences are spliced
Solution Approach 1:
The patent applies parameter changes by introducing optional temporal reference parameters (picture_distance, temporal gap, or temporal difference) that can be included in the bitstream when needed. These parameters modify the HRD operation to account for sequence splicing, allowing the system to maintain simple structure when sequences are not combined while providing enhanced reliability when multiple sequences are concatenated.
Data Source
AI summary
An encoder comprising an input for inputting video signal to be encoded to form an encoded video signal comprising pictures of at least a first coded video sequence and a second coded video sequence, a hypothetical decoder for hypothetically decoding encoded video signal, an encoded picture buffer, and a decoded picture buffer, and a definer for defining a parameter indicative of the temporal difference between the last picture of the first coded video sequence and the first picture of the second coded video sequence in output/display order.