Adaptive Sampling for Graphics Rendering Energy Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional graphics pipelines consume excessive energy due to rendering every region of an image at the same minimum resolution, lacking automation to reduce sampling rates below one sample per pixel, especially in mobile devices.

Innovation Solution

An adaptive sampling method that determines the speed of pixels in each tile, classifying them into speed categories to decide on sampling rates, allowing for reduced sampling in quasi-static or high-speed tiles, with interpolation and advection used to maintain image quality, and dithering to minimize visual artifacts.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If every region of an image is rendered at the same minimum resolution (one sample per pixel), then image quality is maintained, but energy consumption increases

Engineering Contradiction:
Improveimage qualityVSAvoidenergy consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent applies different sampling rates to different spatial regions of the image based on local motion characteristics. Static or slow-moving regions use lower sampling rates (including temporal reuse), while fast-moving regions maintain higher sampling rates, optimizing the balance between image quality and energy consumption on a region-by-region basis

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system dynamically adjusts sampling rates frame-to-frame based on detected motion. The graphics system analyzes motion vectors and pixel displacement between frames to adaptively determine which regions require full sampling and which can use reduced sampling, making real-time decisions about rendering resources

Inventive Principle:
Principle #15Dynamics

2Use of energy by moving object

If sampling rate is reduced below one sample per pixel in local regions, then energy consumption decreases, but image quality may deteriorate

Engineering Contradiction:
Improveenergy consumptionVSAvoidimage quality
Core Design Contradiction:
Use of energy by moving objectVSMeasurement precision

Solution Approach 1:

The system uses motion detection feedback from comparing consecutive frames to make informed sampling decisions. By analyzing pixel displacement and motion vectors, the graphics system receives feedback about which regions actually need high-quality rendering, allowing it to reduce sampling only where motion indicates it won't be noticeable

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The graphics system performs preliminary motion analysis in a preprocessing stage before the actual rendering occurs. By detecting motion and classifying regions in advance, the system can pre-determine which tiles will use reduced sampling and which will use full sampling, avoiding quality issues that would arise from post-processing corrections

Inventive Principle:
Principle #10Preliminary action

3Use of energy by moving object

If motion analysis and adaptive sampling decisions are implemented, then energy efficiency improves, but system complexity increases

Engineering Contradiction:
Improveenergy efficiencyVSAvoidsystem complexity
Core Design Contradiction:
Use of energy by moving objectVSDevice complexity

Solution Approach 1:

The patent divides the image into discrete tiles or blocks that can be independently analyzed and rendered. This segmentation allows motion analysis to be performed on small, manageable regions rather than the entire image, reducing the computational complexity of each decision while maintaining overall energy efficiency benefits

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The graphics system performs its own motion analysis and adaptive sampling decisions autonomously without requiring external control or complex user input. The system services itself by automatically detecting motion, classifying regions, and adjusting sampling rates, which simplifies the overall system architecture despite the added functionality

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP2960863B1Motion based adaptive rendering
Publication Date: 2020.11.11 SAMSUNG ELECTRONICS CO LTD
  • EP2960863B1 patent drawingFigure 1
  • EP2960863B1 patent drawingFigure 2
  • EP2960863B1 patent drawingFigure 3

AI summary

An apparatus, system and method is provided to determine a motion of pixels in local regions of a scene, classify the motion into a speed category, and make decisions on how to render blocks of pixels. In one implementation the motion in a tile is classified into at least three different speed regimes. If the pixels in a tile are in a quasi-static speed regime, a determination is made whether or not to reuse a fraction of pixels from the previous frame. If the pixels are determined to be in a high speed regime, a decision is made whether or not a sampling rate may be reduced.