CABAC Video Bitstream Tranches for Low-Delay Parallel Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current parallel processing techniques in video encoding and decoding, such as those used in HEVC, face challenges in reducing coding delay while maintaining coding efficiency, especially in multi-threaded environments where data locality is compromised, leading to increased end-to-end delays in video transmission.
Innovation Solution
The approach involves segmenting data into smaller tranches within WPP substreams or tiles, allowing for continued context-adaptive binary arithmetic coding probability adaptation across tranche boundaries, enabling earlier transmission and decoding of these tranches, which are smaller than original slices or substreams, thus reducing delay without compromising entropy coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If data is segmented into smaller tranches within WPP substreams or tiles, then transmission speed and decoding start time are improved, but coding efficiency is worsened due to interrupted CABAC probability adaptation
Solution Approach 1:
The patent divides the video data into smaller tranches within WPP substreams or tiles, allowing parallel transmission and decoding. Each tranche can be processed independently by different threads, improving transmission speed and enabling earlier decoding start times without requiring complete substream reception.
Solution Approach 2:
The patent maintains continuous CABAC probability adaptation across tranche boundaries by preserving the context state. This ensures that the entropy coding efficiency is not significantly degraded despite the segmentation, allowing the coding process to continue smoothly across multiple tranches while maintaining compression effectiveness.
2Productivity
If parallel processing is enabled by interrupting spatial dependencies between LCUs, then processing speed is improved, but decoding accuracy is worsened due to loss of contextual information
Solution Approach 1:
The patent segments the picture into tiles and WPP substreams, which are further divided into tranches. This segmentation enables parallel processing by allowing independent decoding of each tranche by different threads, significantly improving processing speed while maintaining sufficient decoding accuracy through careful design of the segmentation boundaries.
Solution Approach 2:
The patent uses context state as an intermediary that carries probabilistic information across tranche boundaries. By preserving and transferring the CABAC context between tranches, the system maintains the necessary contextual information for accurate decoding even though spatial dependencies are interrupted for parallel processing.
3Loss of energy
If complete WPP substreams are transmitted before decoding starts, then coding efficiency is maintained, but end-to-end delay is increased
Solution Approach 1:
The patent divides complete WPP substreams into smaller tranches that can be transmitted and decoded independently. This allows the decoding process to start as soon as the first tranche is received, rather than waiting for the entire substream, significantly reducing end-to-end delay while maintaining acceptable coding efficiency through context state preservation.
Solution Approach 2:
The patent prepares context states in advance and preserves them across tranche boundaries, enabling the decoder to immediately process received tranches without waiting for complete substream reception. This preliminary preparation of context information allows decoding to proceed concurrently with transmission, reducing overall delay.
Data Source
AI summary
A raw byte sequence payload describing a picture in slices, WPP substreams or tiles and coded using context-adaptive binary arithmetic coding is subdivided into tranches with continuing the context-adaptive binary arithmetic coding probability adaptation across tranche boundaries. Thereby, tranche boundaries additionally introduced within slices, WPP substreams or tiles do not lead to a reduction in the entropy coding efficiency of these entities. However, the tranches are smaller than the original slices, WPP substreams or tiles and accordingly they may be transmitted earlier, i.e. with lower delay, than the un-chopped original entities. According to another aspect combinable with the first aspect, substream marker NAL units are used within a sequence of NAL units of a video bitstream to enable a transport demultiplexer to assign data of slices within NAL units to the corresponding substreams or tiles so as to be able to, in parallel, serve a multi-threaded decoder with the corresponding substreams or tiles.