Tile-Based Video Coding for Compact Slice Addressing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video coding technologies face challenges in efficiently compressing video data for transmission and storage, particularly in limited bandwidth environments, with inefficiencies in slice header overhead and processing resources, and the need for improved compression techniques that support parallel processing and reduced memory usage.
Innovation Solution
The method involves partitioning images into slices and tiles, using tile IDs to define slice boundaries, allowing for efficient signaling of slice data without rewriting slice headers, and enabling parallel processing, thereby reducing network and memory resources while increasing decoding speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If video data is compressed using traditional slice-based addressing, then the video data can be transmitted and stored, but the bitstream size is larger and requires more processing resources
Solution Approach 1:
The image is divided into multiple slices, and each slice is further divided into tiles with unique tile IDs. This hierarchical segmentation allows for more efficient addressing and signaling, reducing bitstream overhead while enabling parallel processing of different tile regions.
Solution Approach 2:
The patent introduces a new dimension for slice addressing by using tile ID relationships (top-left and bottom-right corner tiles) instead of traditional linear or relative positioning methods. This dimensional change in addressing enables more compact representation of slice boundaries and supports parallel decoding operations.
2Measurement precision
If slice headers are rewritten for sub-frame signaling, then accurate addressing is achieved, but processing resources and time are consumed
Solution Approach 1:
Tile IDs are assigned and established in advance during the encoding phase, creating a predefined addressing framework. This preliminary action eliminates the need for runtime header rewriting, as the tile ID-based addressing system is already in place and can be directly used for sub-frame signaling and slice identification.
3Loss of information
If all tile IDs are explicitly signaled in the bitstream, then complete information is provided, but the bitstream becomes larger
Solution Approach 1:
The patent extracts only the essential tile information (top-left and bottom-right corner tile IDs of each slice) from the complete set of tile IDs. This selective extraction maintains sufficient information for accurate slice boundary determination and parallel processing while significantly reducing the amount of data that needs to be transmitted in the bitstream.
Solution Approach 2:
The decoder is enabled to infer intermediate tile IDs within a slice by using the top-left and bottom-right corner tile IDs as boundaries. This self-service mechanism allows the decoder to reconstruct the complete tile ID sequence without it being explicitly transmitted, reducing bitstream overhead while maintaining information completeness.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A video coding mechanism is disclosed. The mechanism includes receiving a bitstream including image data coded in a plurality of slices. A top left tile identifier (ID) and a bottom right tile ID of a first slice are determined. Boundaries of the first slice are determined based on the top left tile ID and the bottom right tile ID. The first slice is decoded to generate a reconstructed image based on the boundaries of the first slice.