Video Tile Map for Parallel Bitstream Decomposition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding standards, such as HEVC, are not designed to support high-level parallelisation architectures, making it difficult to decompose a bitstream into separate bitstreams for processing by multiple decoder instances, especially for high-resolution videos like 8K.
Innovation Solution
The proposed solution involves organizing encoded video data in video tiles and using tile positioning information, including tile identifiers and boundary identifiers, to create a tile map that defines the spatial layout of tiles in an output video frame. This allows for flexible bitstream decomposition and parallel processing of video tiles by different decoder instances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If video coding standards like HEVC are used, then video compression efficiency is improved, but the ability to decompose bitstream into separate bitstreams for parallel processing deteriorates
Solution Approach 1:
The patent divides the video bitstream into multiple independently decodable bitstream parts, each corresponding to a specific video tile. This segmentation allows parallel processing while maintaining the compression efficiency of HEVC, as each tile can be decoded independently without requiring other tiles.
Solution Approach 2:
The patent introduces a new dimension of organization by creating a tile-based spatial structure with explicit positioning information. This allows the bitstream to be decomposed along spatial dimensions into multiple parallel streams while preserving the temporal and spatial compression benefits of HEVC.
2Productivity
If bitstream decomposition into multiple bitstreams is implemented, then parallel processing capability is improved, but bitstream structure complexity increases
Solution Approach 1:
The bitstream is segmented into multiple tile-based parts with clear boundaries and positioning information. Each segment is self-contained and can be processed independently, enabling parallel decoding while maintaining a relatively simple overall structure through systematic organization.
Solution Approach 2:
The patent introduces tile positioning information as an intermediary element that facilitates the decomposition process. This intermediary structure provides the necessary spatial context for parallel processing without requiring complex inter-dependencies between bitstream parts.
3Productivity
If tile-based decomposition is implemented, then decoding efficiency for high-resolution videos is improved, but computational resources required for tile management increase
Solution Approach 1:
By segmenting high-resolution video into smaller tile-based units, the patent enables parallel decoding that distributes computational load across multiple processing units. This segmentation improves overall decoding efficiency while the systematic tile management minimizes overhead.
Solution Approach 2:
The patent performs preliminary organization of video data into tiles with explicit positioning information during encoding. This preliminary action reduces the computational burden during decoding, as the spatial structure is already established and does not require complex analysis during the decoding phase.
Data Source
AI summary
A method is described for forming output video frames by a decoder device, comprising: receiving a bitstream comprising bitstream parts representing encoded media data of video tiles and tile positioning information associated with the encoded video tiles; extracting tile positioning information from the bitstream, the tile positioning information comprising tile identifiers for identifying encoded video tiles, a tile identifier being associated with one or more boundary identifiers, each identifying a boundary of a video tile identified by a tile identifier; determining a tile map based on the boundary identifiers, the tile map representing a spatial layout of video tiles in an output video frame; and forming an output video frame, including identifying parts of the bitstream representing encoded media data of video tiles identified in the tile map, decoding the identified encoded media data and copying the decoded media data in the output video frame buffer according to spatial layout.