Scalable Video Coding SPS Reuse for Reduced Redundancy

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional video coding methods, such as those in the H.264/AVC scalable extension, face inefficiencies in data transmission due to redundant sequence parameter sets (SPSs) and limited flexibility, leading to increased session setup delays and reduced coding efficiency.

Innovation Solution

The method allows for the reuse of a sequence parameter set across layers with different dependency identifiers, activating a new SPS only when necessary, and using the same SPS for base and enhancement layers with equal parameters, thereby reducing redundant data transmission and enhancing flexibility.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Use of energy by moving object

If conventional video coding standards (H.264, H.265) are used, then video quality can be maintained, but computing power consumption increases exponentially with resolution and frame rate

Engineering Contradiction:
Improvecomputing power consumptionVSAvoidvideo coding efficiency
Core Design Contradiction:
Use of energy by moving objectVSProductivity

Solution Approach 1:

The patent transforms the video coding problem from spatial domain to frequency domain by applying 2D Discrete Cosine Transform (DCT) to convert spatial samples into frequency coefficients. This parameter transformation enables more efficient energy compaction and coding, reducing computing power consumption while maintaining video quality. The frequency domain representation allows for better exploitation of video data characteristics and more effective compression.

Inventive Principle:
Principle #35Parameter changes

2Manufacturing precision

If high resolution and frame rate are used, then video quality improves, but computing power requirements increase exponentially

Engineering Contradiction:
Improvevideo qualityVSAvoidcomputing power requirements
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent replaces traditional complex spatial domain coding mechanisms with frequency domain transformation (2D-DCT). This substitution enables more efficient representation of video data by exploiting frequency characteristics, thereby reducing the computational complexity required to achieve high video quality at high resolutions and frame rates.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If conventional coding methods are used for 8K and 4K videos, then video data can be processed, but the computational complexity becomes unsustainable

Engineering Contradiction:
Improvevideo processing capabilityVSAvoidcomputational complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent applies 2D Discrete Cosine Transform to convert spatial video data into frequency domain coefficients, fundamentally changing the parameter representation. This transformation enables more efficient processing of high-resolution video data by compacting energy into fewer coefficients, thereby reducing computational complexity while maintaining or improving processing capability for 8K and 4K videos.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP1869888B1Method, device and system for effectively coding and decoding of video data
Publication Date: 2016.07.06 NOKIA TECHNOLOGIES OY
  • EP1869888B1 patent drawingFigure 1
  • EP1869888B1 patent drawingFigure 2
  • EP1869888B1 patent drawingFigure 3

AI summary

The present invention discloses methods and devices for video encoding and decoding, and more specifically to scalable video data processing, A method for scalable encoding of video data is provided. According to a first operation obtaining of said video data is provided. Then obtaining said video data is carried out, followed by generating a base layer based on said obtained video data, the base layer comprising at least one picture, generating at least one enhancement layer based on said obtained video data, the at least one enhancement layer comprising at least one picture, generating a dependency identifier (DependencyID) for each of said base and enhancement layers, each DependencyID being associated with a reference number; determining a respective sequence parameter set (SPS) for each of said base layer and said at least one enhancement layer having different DependencyID values, wherein for a number of base and enhancement layers having SPS parameters substantially ihe same, using one SPS; and encoding said base layer and said at least one enhancement layer by using determined SPS 's.