Parallel Decodable Slices for Multi-View Video Coding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multi-view video decoding in 3D-TV systems is hindered by inter-view dependencies, requiring sequential decoding of views, which leads to complexity and inefficiency, especially as the number of views increases, necessitating decoders to run significantly faster than regular video decoders.
Innovation Solution
Implementing parallel decoder architecture by encoding views with constraints that allow macroblocks in one view to depend only on reconstruction values of a subset of macroblocks in other views, with signaling of these constraints in the bitstream, enabling simultaneous decoding of macroblocks across views.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If inter-view dependencies are exploited for multi-view video coding, then coding efficiency is improved, but decoding parallelism deteriorates requiring sequential decoding
Solution Approach 1:
The video stream is segmented into independent slices within each view, where each slice can be decoded independently. This segmentation allows parallel decoding of different slices across multiple views simultaneously, while still maintaining inter-view dependencies within each slice for coding efficiency.
Solution Approach 2:
The patent introduces a new dimension of parallelism by organizing decoding in a two-dimensional grid of views and slices. Instead of sequential decoding across views, the system enables parallel decoding across both view dimension and slice dimension, transforming the decoding architecture from purely sequential to multi-dimensional parallel processing.
2Manufacturing precision
If sequential decoding of views is implemented to maintain inter-view dependencies, then decoding accuracy is improved, but output delay increases
Solution Approach 1:
The patent performs preliminary actions by pre-processing and organizing reference data from dependent views before actual decoding begins. Reference frames and slices are prepared in advance, allowing the decoder to start processing multiple views in parallel without waiting for sequential completion of previous views, thus reducing output delay while maintaining accuracy.
Solution Approach 2:
The decoding system dynamically adjusts the processing order and timing of different views based on slice availability and inter-view dependency requirements. Rather than rigid sequential processing, the system flexibly schedules decoding operations to maximize parallelism while ensuring that required reference data is available when needed, reducing overall output delay.
3Adaptability or versatility
If more views are displayed simultaneously in 3D-TV systems, then user experience is improved, but decoder speed requirement increases significantly
Solution Approach 1:
Each view is divided into multiple independent slices that can be decoded in parallel. This segmentation allows the decoder to process multiple views simultaneously by working on different slices of different views in parallel, rather than sequentially processing entire views, thus supporting more views without proportionally increasing overall decoder speed requirements.
Solution Approach 2:
The patent merges the processing of multiple views by sharing common reference data and using unified decoding logic across views. By combining views into a coordinated parallel processing framework where slices from different views can be processed together, the system efficiently handles a larger number of views without requiring each view to be decoded at full independent speed.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system and method for enabling parallel decoder implementation for different views, even when there are existing dependencies between views. In various embodiments of the present invention, information is signaled to a decoder that slices are coded using certain constraints, so that parallel decoding of slices is possible. This signaling can be performed at a sequence parameter set level, or it could be performed at picture parameter set level, slice header level or macroblock header level. Additionally, the delay between various views is also signaled to the decoder, at the same location as the constraint information in various embodiments. Various algorithms can also be used to improve the coding efficiency of the system.