Pipelined Image Processing Engine Block Segmentation Latency

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional pipelined image processing systems suffer from high latency and reduced scalability due to limitations in distributing hardware resources among multiple image processing effects, leading to poor throughput and increased complexity, especially when chaining discrete effects that depend on consecutive execution.

Innovation Solution

The implementation of a flexible and scalable pipeline that breaks down image frames into blocks, allowing each effect to process blocks independently and distribute them across multiple processing nodes, with dynamic load balancing to optimize resource allocation and reduce latency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If discrete effects are chained together in a pipelined system for image processing, then processing capacity is increased, but latency increases correlating with the number of effects in the pipeline

Engineering Contradiction:
Improveprocessing capacityVSAvoidpipeline latency
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent segments image frames into smaller blocks that can be processed independently through the effect pipeline. This allows multiple blocks to be processed in parallel across different pipeline stages, reducing the overall latency while maintaining high processing capacity. Each block can traverse the pipeline independently, eliminating the sequential bottleneck of traditional frame-based processing.

Inventive Principle:
Principle #1Segmentation

2Productivity

If multiple discrete effects are chained in consecutive execution, then processing capability is enhanced, but system complexity increases reducing extensibility and scalability

Engineering Contradiction:
Improveprocessing capabilityVSAvoidsystem complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

By dividing the image processing into independent block-level operations, the system achieves modular architecture where effects can be added, removed, or modified without affecting the overall pipeline structure. This segmentation enables scalable deployment from single-core to multi-core systems.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent creates a universal pipeline architecture that can handle multiple types of image processing effects (spatial, temporal, color space conversions) through a common block processing framework. This multi-functional design allows the same infrastructure to support diverse effects, reducing system complexity while enhancing processing capability.

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

3Ease of operation

If frame-based processing is used where each effect operates on complete frames, then independence among effects is maintained, but throughput decreases due to high latency

Engineering Contradiction:
Improveeffect independenceVSAvoidthroughput
Core Design Contradiction:
Ease of operationVSProductivity

Solution Approach 1:

The patent segments complete frames into smaller blocks that maintain independence for processing while enabling parallel execution. Each block can be processed independently through the pipeline, preserving effect independence while dramatically improving throughput by allowing multiple blocks to be in different pipeline stages simultaneously.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary segmentation of frames into blocks before processing, preparing the data structure to enable parallel pipeline execution. This preliminary action allows the pipeline to process multiple blocks concurrently, increasing throughput while maintaining the independence of individual effect operations.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2161685B1Pipelined image processing engine
Publication Date: 2018.11.28 SONY GROUP CORP
  • EP2161685B1 patent drawingFigure 1
  • EP2161685B1 patent drawingFigure 2
  • EP2161685B1 patent drawingFigure 3

AI summary

A method of processing image frames through a pipeline of effects includes breaking the image frames into multiple blocks of image data. The example method includes generating a plurality of blocks from each frame, processing each block through a pipeline of effects in a predefined consecutive order, and aggregating the processed blocks to produce an output frame by combining the primary pixels from each processed block. The pipeline of effects may be distributed over a plurality of processing nodes, and each effect may process a block, provided as input to the node. Each processing node may independently process a block using an effect.