Video Accelerator Adaptive Inter-Prediction Mode Selection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Video accelerators face challenges in processing video data due to limited internal memory and bandwidth constraints, leading to delays in the coding process when handling multiple video streams simultaneously, as they need to allocate bandwidth among streams and wait for sufficient memory bandwidth to read reference data for inter-prediction modes.

Innovation Solution

The video accelerator adaptively selects inter-prediction modes and reduces the size of reference windows based on available bandwidth, enabling modes that use less bandwidth and perform predictions without waiting for sufficient memory access, thereby maintaining output data rates even if optimal compression is not achieved.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If the video accelerator uses inter-prediction modes with larger reference windows to achieve optimal compression, then manufacturing precision is improved, but productivity deteriorates due to bandwidth limitations and processing delays

Engineering Contradiction:

Improvecompression qualityVSAvoidoutput data rate

Core Design Contradiction:

Manufacturing precisionVSProductivity

Solution Approach 1:

The patent implements dynamic selection of prediction modes and reference window sizes based on real-time bandwidth availability. The system adapts between different inter-prediction modes (e.g., switching between large reference windows for high bandwidth conditions and smaller windows or alternative modes for limited bandwidth), allowing the video accelerator to optimize compression quality when resources permit while maintaining productivity when bandwidth is constrained

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes operational parameters (prediction mode selection, reference window size) based on available bandwidth conditions. By monitoring bandwidth availability and adjusting these parameters dynamically, the system resolves the contradiction between achieving optimal compression (which requires larger reference windows) and maintaining high output data rates (which requires faster processing with smaller windows or fewer memory accesses)

Inventive Principle:

Principle #35Parameter changes

2Manufacturing precision

If the video accelerator waits for sufficient memory bandwidth to read reference data, then manufacturing precision is improved by using optimal inter-prediction modes, but loss of time increases causing processing delays

Engineering Contradiction:

Improveprediction accuracyVSAvoidprocessing delay

Core Design Contradiction:

Manufacturing precisionVSLoss of time

Solution Approach 1:

The system dynamically adjusts its operation based on real-time bandwidth conditions. When bandwidth is sufficient, it uses optimal inter-prediction modes with larger reference windows for better accuracy. When bandwidth is limited, it switches to faster alternatives that require fewer memory accesses, thus eliminating the need to wait for bandwidth availability and reducing processing delays

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system performs preliminary assessment of bandwidth availability before selecting prediction modes. By determining whether sufficient bandwidth is available in advance, the system can proactively choose the appropriate processing path - either preparing to use optimal modes with larger reference windows or switching to faster modes that don't require waiting for bandwidth, thereby preventing delays before they occur

Inventive Principle:

Principle #10Preliminary action

3Productivity

If the video accelerator processes multiple video streams simultaneously to increase productivity, then output data rate is improved, but device complexity increases due to bandwidth allocation requirements

Engineering Contradiction:

Improveoutput data rateVSAvoidbandwidth management complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The video accelerator is designed with multi-functionality to handle multiple video streams simultaneously. By implementing a unified bandwidth management mechanism that can dynamically allocate resources across multiple streams and adapt prediction modes based on overall system bandwidth availability, the system achieves high productivity for multiple streams without proportionally increasing complexity

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system uses parameter changes in bandwidth allocation and prediction mode selection to manage multiple streams efficiently. By monitoring overall bandwidth usage and dynamically adjusting allocation parameters across streams, the system maintains manageable complexity while achieving high aggregate output data rates

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10694189B1Adaptive use of search modes based on bandwidth availability

Publication Date: 2020.06.23 AMAZON TECH INC

AI summary

Reference data is one type of data that the video accelerator may frequently be read from external memory. In various examples, the video accelerator can adaptively select inter-prediction modes based on the bandwidth to external memory that is available at any point in time. The video accelerator can determine the amount of bandwidth that is available, and when the bandwidth is insufficient for obtaining reference data for all possible inter-prediction modes, the video accelerator can select an inter-prediction mode based on the size of the reference window associated with the inter-prediction mode, the size being within an amount of data that can be read with the available bandwidth. The video accelerator can then obtain a reference window from external memory, and perform prediction using the selected inter-prediction mode and the reference window.