Video Encoding Tile Slice Partitioning Parallel Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video compression techniques face challenges in efficiently compressing high-resolution and high-quality images, leading to increased data transfer and storage costs. There is a need for improved methods to define process units in pictures, impose constraints on encoding/decoding, and enable smooth parallel processing.
Innovation Solution
The proposed solution involves partitioning an input picture into tiles and slices, applying constraints for parallel processing, and using non-square quad-tree transform (NSQT) and asymmetric motion partition (AMP) techniques. The video encoder imposes constraints on the use of NSQT and AMP and transmits indicators, while the decoder uses these techniques accordingly to enhance compression efficiency and reduce complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If high-resolution and high-quality image compression is performed, then image quality is improved, but data transfer cost and storage cost increase
Solution Approach 1:
The picture is divided into multiple tiles and slices, allowing independent processing and parallel decoding. This segmentation enables efficient compression of high-resolution images by processing smaller units independently, reducing the overall data amount while maintaining high image quality through targeted compression techniques applied to each segment.
Solution Approach 2:
Asymmetric motion partition (AMP) divides motion compensation blocks into asymmetric sub-blocks rather than uniform squares. This allows more efficient representation of motion patterns in high-quality images, reducing the data amount needed to maintain manufacturing precision by adapting the partitioning to actual motion characteristics.
2Productivity
If parallel processing is enabled for video decoding, then processing speed is improved, but complexity of managing process units increases
Solution Approach 1:
The video picture is segmented into tiles and slices that can be decoded in parallel. This segmentation structure enables multiple processing units to work simultaneously on different portions of the picture, improving processing speed while the hierarchical organization (tiles containing slices) manages the complexity of coordinating these parallel operations.
Solution Approach 2:
The picture is divided not only horizontally into slices but also vertically into tiles, creating a two-dimensional grid structure for parallel processing. This multi-dimensional segmentation allows for flexible parallel processing configurations, improving productivity while distributing the management complexity across multiple independent dimensions.
3Productivity
If non-square quad-tree transform and asymmetric motion partition are used, then compression efficiency is improved, but constraint management complexity increases
Solution Approach 1:
Asymmetric motion partition divides motion blocks into non-uniform sub-blocks that better match actual motion patterns, improving compression efficiency. The patent manages the resulting complexity by establishing systematic constraints on AMP usage and integrating it with the tile-slice partitioning structure, allowing efficient compression without overwhelming constraint management overhead.
Solution Approach 2:
The transform and partition structures are made adaptive rather than fixed. Non-square quad-tree transform allows transform block sizes to vary dynamically based on content characteristics, improving compression efficiency. The system manages complexity through controlled constraints that allow dynamic adaptation within bounded parameters.
Data Source
AI summary
The present invention relates to a video encoding method, a video decoding method, and a device using the same, and the video encoding method according to the present invention comprises the steps of: specifying a tile and a slice by partitioning an inputted picture; performing encoding on the basis of the tile and the slice; and transmitting the encoded video information, wherein the picture is partitioned into one or more tiles and one or more slices, and the restrictions for parallel processing can be applied to the tiles and the slices.