Parallel Video Encoders with Shared Reference Frame Cache

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Contemporary video encoding systems face challenges in handling high bandwidth requirements due to increased frame rates, particularly in space-limited systems, where adding more DRAM may not be viable, and existing encoding standards like H.264 struggle to efficiently manage the increased video data traffic without significant bandwidth increases.

Innovation Solution

A video encoding apparatus with a reference frame cache and parallel video encoders that share the same cached reference frame data, allowing for encoding operations to be performed in parallel across multiple source frames, thereby reducing bandwidth requirements by using the same reference frame data for multiple encoding operations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If the frame rate is increased to meet consumer demand for higher specifications and features, then the video quality and functionality are improved, but the system bandwidth requirement increases proportionally

Engineering Contradiction:
Improveframe rateVSAvoidbandwidth
Core Design Contradiction:
ProductivityVSQuantity of substance

Solution Approach 1:

Multiple video encoders share a common reference frame cache, merging the reference frame storage resource. This allows multiple encoding operations to access the same reference frame data simultaneously, reducing the total bandwidth required to supply reference frames to multiple encoders and enabling higher frame rates without proportional bandwidth increases.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent transitions from sequential encoding of frames to parallel encoding across multiple frames by utilizing a shared reference frame cache. This dimensional change in processing architecture allows multiple source frames to be encoded simultaneously while reusing the same reference frame data, decoupling the frame rate increase from proportional bandwidth increases.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Quantity of substance

If additional DRAM devices are added to increase bandwidth capacity, then the system can handle higher video data traffic, but the device complexity and space requirements increase

Engineering Contradiction:
ImprovebandwidthVSAvoidsystem complexity
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The reference frame cache serves multiple functions: it stores reference frames for multiple source frames simultaneously, supports multiple video encoders, and enables parallel encoding operations. This multi-functional design eliminates the need for separate reference frame storage for each encoder, reducing overall system complexity and DRAM requirements while maintaining high bandwidth capacity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Quantity of substance

If additional DRAM devices are added to increase bandwidth capacity, then the system can handle higher video data traffic, but the space requirements increase which may not be viable in space-limited systems

Engineering Contradiction:
ImprovebandwidthVSAvoidspace
Core Design Contradiction:
Quantity of substanceVSVolume of moving object

Solution Approach 1:

Multiple video encoders share a common reference frame cache, merging the reference frame storage resource. This allows multiple encoding operations to access the same reference frame data simultaneously, reducing the total bandwidth required to supply reference frames to multiple encoders and enabling higher frame rates without proportional bandwidth increases.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The reference frame cache serves multiple functions: it stores reference frames for multiple source frames simultaneously, supports multiple video encoders, and enables parallel encoding operations. This multi-functional design eliminates the need for separate reference frame storage for each encoder, reducing overall system complexity and DRAM requirements while maintaining high bandwidth capacity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS9591319B2Video encoder
Publication Date: 2017.03.07 ARM LTD
  • US9591319B2 patent drawing
  • US9591319B2 patent drawing
  • US9591319B2 patent drawing

AI summary

A video encoding apparatus for encoding a video stream comprises a reference frame cache for reference frame video data retrieved from a reference frame storage unit in external memory, which is derived from an individual frame of the video stream. First and second source frame storage units store first and second blocks of unencoded video data taken from first and second source frames of the video stream, respectively. First and second video encoders perform first and second encoding operations to encode the first and second blocks of unencoded video data with reference to the reference frame video data cached in the reference frame cache, respectively. The first video encoder and the second video encoder perform the first encoding operation and the second encoding operation in parallel with one another.