Universal Video Transcoding System Using Motion Vector Mapping

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing video transcoding architectures are limited in their ability to perform multiple transcoding operations efficiently, as they are typically designed for specific use cases and lack flexibility, leading to high computational complexity and increased software development and maintenance costs.

Innovation Solution

A computerized method and system that reads multiple transcoding use cases, decodes input images, maps motion vectors and macroblock modes, and re-encodes images to adapt to various formats, allowing for one-to-one and one-to-many mappings of motion vectors and macroblock modes to perform multiple transcoding operations efficiently.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If existing video transcoding architectures are designed for specific use cases, then they can achieve good performance for that particular use case, but they lack flexibility and cannot perform multiple transcoding operations efficiently

Engineering Contradiction:

Improveability to perform multiple transcoding operationsVSAvoidcomputational complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements a universal transcoding architecture that can perform multiple transcoding operations (bit rate adaptation, spatial resolution adaptation, temporal resolution adaptation, compression format conversion) using a single unified system. The architecture uses a common decoding module, motion vector mapping module, and re-encoding module that can handle different transcoding use cases by selecting appropriate processing paths, eliminating the need for separate dedicated systems for each transcoding task.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent divides the transcoding process into distinct functional segments: decoding module, motion vector mapping module, and re-encoding module. Each segment can independently process different aspects of transcoding operations, allowing the system to handle multiple use cases by combining different segments in various configurations without increasing overall system complexity.

Inventive Principle:

Principle #1Segmentation

2Adaptability or versatility

If existing video transcoding architectures are designed for specific use cases, then they can optimize for that use case, but software development and maintenance costs increase

Engineering Contradiction:

Improveflexibility to support multiple use casesVSAvoidsoftware development and maintenance costs

Core Design Contradiction:

Adaptability or versatilityVSEase of manufacture

Solution Approach 1:

The patent creates a single unified software architecture that can support multiple transcoding use cases through a common set of modules. The decoding module, motion vector mapping module, and re-encoding module serve all transcoding operations, reducing the total codebase size and eliminating the need to maintain multiple separate software systems for different transcoding tasks.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent implements a dynamic configuration system where the motion vector mapping module can adaptively select different mapping strategies (one-to-one, one-to-many, many-to-one, many-to-many) based on the specific transcoding use case. This dynamic approach allows a single software system to optimize performance for different operations without requiring separate software versions or modules for each use case.

Inventive Principle:

Principle #15Dynamics

3Adaptability or versatility

If multiple transcoding operations are performed using conventional methods, then various adaptations can be achieved, but computational complexity increases significantly

Engineering Contradiction:

Improvesupport for multiple transcoding use casesVSAvoidcomputational complexity

Core Design Contradiction:

Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The patent merges multiple transcoding operations into a single integrated processing pipeline. The decoding module, motion vector mapping module, and re-encoding module work together in sequence to perform multiple adaptations (bit rate, spatial resolution, temporal resolution, compression format) in one unified operation, avoiding the need to run separate transcoding processes that would multiply computational complexity.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The patent performs preliminary decoding of the input video stream and extraction of motion vectors before the actual transcoding operations. This preliminary action prepares the data in advance, allowing the subsequent motion vector mapping and re-encoding steps to proceed more efficiently without needing to re-decode or re-process the entire stream for each different transcoding use case.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS9420284B2Method and system for selectively performing multiple video transcoding operations

Publication Date: 2016.08.16 ECOLE DE TECH SUPERIEURE

AI summary

Method and system for transcoding a sequence of input images into a sequence of output images that can effectively handle one or more transcoding uses cases are presented. The embodiments of the invention exploit the incoming metadata retrieved from the decoding process to adapt the video content and to achieve a significant speed-up in comparison to the traditional cascaded approach while maintaining high quality for output images.