Video Proxy Codec Pre-Processor for Compression Artifact Compensation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing video codecs introduce visually undesirable artifacts during compression, and new standards take years to develop, necessitating a solution to improve compression quality without changing the codec's standard.

Innovation Solution

A pre-processor is trained using a differentiable proxy codec to compensate for distortion, allowing it to enhance the quality of decoded content while maintaining the original bitrate, by adding alterations to the source content before encoding.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If existing codecs are used to compress content, then the compression process is fast and compatible with current standards, but visually undesirable artifacts are introduced and compression quality deteriorates

Engineering Contradiction:

Improvecompression qualityVSAvoidcompression artifacts

Core Design Contradiction:

Manufacturing precisionVSObject-generated harmful factors

Solution Approach 1:

The pre-processor applies modifications to the source video content before encoding by the existing codec. This preliminary action compensates for expected compression artifacts, allowing the content to maintain higher quality after compression without changing the codec standard itself.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system embraces the inevitability of compression artifacts and uses machine learning models to predict and counteract them. By training the pre-processor on compressed content, it learns to add compensatory information that becomes beneficial after the codec introduces its expected distortions.

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

2Manufacturing precision

If new codec standards are developed to improve compression quality, then compression performance improves, but development time increases and implementation is delayed

Engineering Contradiction:

Improvecompression qualityVSAvoidstandard development time

Core Design Contradiction:

Manufacturing precisionVSLoss of time

Solution Approach 1:

The pre-processor acts as an intermediary layer between the source content and the existing codec. It prepares the content in advance to compensate for codec limitations, effectively bridging the quality gap without requiring new codec standards or lengthy development cycles.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

Rather than waiting for new standards to be developed and implemented, the system performs preliminary processing of the video content to pre-compensate for compression artifacts. This allows immediate quality improvement without the time investment required for standard development.

Inventive Principle:

Principle #10Preliminary action

3Manufacturing precision

If a pre-processor is trained using a differentiable proxy codec to compensate for distortion, then the quality of decoded content improves, but the system complexity increases

Engineering Contradiction:

Improvedecoded content qualityVSAvoidsystem complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent replaces traditional mechanical or algorithmic video processing methods with a machine learning-based pre-processor. The differentiable proxy codec and neural network model substitute for conventional compression optimization approaches, enabling automatic learning of compensation strategies without manual tuning of complex parameters.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS20250220197A1Performance optimization of pre-processor using video proxy codec

Publication Date: 2025.07.03 DISNEY ENTERPRISES INC

AI summary

In some embodiments, a method receives source content. A pre-processor pre-processes the source content to output pre-processed source content. The pre-processor includes a first parameter that is trained based on a differentiable proxy codec, and a calculated adjustment to a second parameter of the differentiable proxy codec is used to train the first parameter of the pre-processor. The method encodes the pre-processed source content into compressed pre-processed source content. The compressed pre-processed source content is output.