Power-Saving Techniques with High Visual-Quality for Mobile

Power-Saving Techniques with High Visual-Quality for Mobile Power-Saving Techniques with High Visual-Quality for Mobile - Start

Added : 2015-12-09 Views :83K

Download Presentation

Power-Saving Techniques with High Visual-Quality for Mobile




Download Presentation - The PPT/PDF document "Power-Saving Techniques with High Visual..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.



Presentations text content in Power-Saving Techniques with High Visual-Quality for Mobile

Slide1

Power-Saving Techniques with High Visual-Quality for Mobile Displays

Dep. of Computer Science & Engineering

Yuan

Ze

University

Speaker: Chun-Han Lin

National Taiwan Normal University

Slide2

Outline

IntroductionLiquid Crystal DisplaysOrganic Light-Emitting Diode DisplaysConclusion

Chun-Han Lin, NTNU

Slide3

Motivation

Mobile applications and services are having a profound effect on people's lifestylesThe energy consumption of mobile devices is a major challenge in sustaining the applications and services

Chun-Han Lin, NTNU

Slide4

Possible Solution

Battery Extenders

Power-Saving Techniques

Chun-Han Lin, NTNU

Slide5

Power Consumption

The display subsystem stays in active mode for various applicationsLiquid Crystal Displays (LCDs)Organic Light-Emitting Diode (OLED) Displays

Chun-Han Lin, NTNU

Slide6

Outline

IntroductionLiquid Crystal DisplaysOrganic Light-Emitting Diode DisplaysConclusion

Chun-Han Lin, NTNU

Slide7

Thin-Film Transistor LCDs

Chun-Han Lin, NTNU

Slide8

Mobile LCDs

What hardware to target?

Chun-Han Lin, NTNU

Power distribution on HTC Desire when browsing videos on YouTube

Power distribution on Apple iPad when browsing videos YouTube

Slide9

LCD Power-Saving Techniques

Dim the backlightImage distortionChallengeLimit the distortionImage compensation techniques

Chun-Han Lin, NTNU

Slide10

Video

A video stream comprises a series of image framesChallengeFlickering effectsInterframe brightness distortionHardware requires time to react and adjust the backlightPrevious workGroups the image frames of a videoQuantizes the number of backlight levelsAdjacent frames, instead of having an overall consideration based on all the frames

Chun-Han Lin, NTNU

Slide11

Backlight scalingDynamically adjust backlight levels for video framesVideo distortionHardware/software limitationUser perceptionEtc.

Backlight Scaling Technique

Chun-Han Lin, NTNU

Slide12

Input and Output

Input dataVideoConstraintVideo distortionHardware/software limitationUser perceptionPower model of mobile deviceOutput dataBacklight file

Chun-Han Lin, NTNU

Slide13

Flowchart

Image frames in video

Backlight assignment

Chun-Han Lin, NTNU

Slide14

Algorithm 1

ChallengeVideo distortionUser perceptionSolutionAvoid abrupt changes in backlight levels

Chun-Han Lin, NTNU

Slide15

Principle of A1

Chun-Han Lin, NTNU

Slide16

Algorithm 2

ChallengeVideo distortionHW/SW limitationSolutionAvoid frequent changes in the backlight level

Chun-Han Lin, NTNU

Slide17

Dynamic-Programming in A2

Min. Energy

Min. Energy

Chun-Han Lin, NTNU

Slide18

Principle of A2

Chun-Han Lin, NTNU

Slide19

Algorithm 3

ChallengeVideo distortionUser perceptionHW/SW limitationSolutionAvoid abrupt changesAvoid frequent changes

Chun-Han Lin, NTNU

Slide20

Dynamic-Programming in A3

Chun-Han Lin, NTNU

Slide21

Principle of A3

Chun-Han Lin, NTNU

Slide22

Cloud-Based Power-Saving Services

With the service is applied, the service provider help reduce the energy consumption of mobile devices when they access Internet applications

Chun-Han Lin, NTNU

Slide23

System Architecture

Video Stream

Backlight File

Backlight Server

Streaming Server

Mobile Device

Chun-Han Lin, NTNU

Slide24

Responsible for generating backlight files

The Cloud Side

Phase 1

Analyze

the

video to decide the

critical backlight levels

(i.e., the dimmest backlight level with respect to the tolerable video distortion)

Phase 2

Determine an optimal backlight assignment for the video based on the device’s power model and capability

Backlight File

Download the requested video from YouTube

Critical backlight levels of the video

Chun-Han Lin, NTNU

Slide25

The Device Side

Measure

power modelsDevelop mobile application programs

iPad’s

display subsystem

Power Monitors

Chun-Han Lin, NTNU

Slide26

Demonstration

Approach validation

Performance evaluation

Chun-Han Lin, NTNU

Slide27

System Deployment

Case studies

System architecture

Chun-Han Lin, NTNU

Slide28

Backlight File

Process Time & Transmission Delay

Approach Validation

Video Download

Phase IPhase II124 seconds1020 seconds2.1 seconds

Transmission

Delay 335 millisecond

s

Cloud Side

Device Side

Chun-Han Lin, NTNU

Slide29

Performance Evaluation

Experimental Results

Case Studies

Chun-Han Lin, NTNU

Slide30

Outline

IntroductionLiquid Crystal DisplaysOrganic Light-Emitting Diode DisplaysConclusion

Chun-Han Lin, NTNU

Slide31

Organic Light-Emitting Diode (OLED) Displays

Chun-Han Lin, NTNU

Slide32

Mobile OLED Displays

OLED is deemed promising technology to replace LCD for mobile displaysBrighter colors, wider viewing angles, faster response times, etc.Power consumption increases dramatically with the pixel values of the displayed image

Chun-Han Lin, NTNU

Slide33

Low-Power Techniques for OLED Displays

Partial display disabling or dimmingDarken the contents that are not of interestImpact user perceptionColor remappingChange colors into colors that consume less powerSuit for GUI but not natural imagesOLED dynamic voltage scalingDecrease the supply voltage of each pixel’s circuitRequire hardware support and partition the display into rectangular regions

Chun-Han Lin, NTNU

Slide34

Inspired by Human Visual Attention

Different regions in an image Receive varying degrees of visual attentionCan tolerate different degrees of image distortion

Chun-Han Lin, NTNU

Slide35

Quality-Retaining Power Saving Technique

Image pixel scalingSegmentationScalingCombination

Chun-Han Lin, NTNU

Slide36

Fast and optimal withoutaccurate OLED power models

Distortion (SSIM)

Analysis

Attention (

Itti

)

Perception (JND)

Conversion

Optimal Algorithm

Chun-Han Lin, NTNU

Slide37

Visual Attention

Not every region in an image receives the same attention levelImage can be segmented based on its saliency map into a set of attention regionsThe saliency map indicates a saliency value for each pixel in an image

Chun-Han Lin, NTNU

Slide38

Image Distortion

Different regions in an image receive varying degrees of attentionDifferent regions can tolerate different degrees of image distortion.Attention regions should be given tolerable distortion in inverse proportion to their attention levels

Chun-Han Lin, NTNU

Slide39

Perception

Lowering the pixel values by applying the critical scaling ratio to each region may result in sharp edges between adjacent regionsThese sharp edges will severely interfere with visual experienceThe difference between the scaling ratios applied to two adjacent regions should be limited

Chun-Han Lin, NTNU

Slide40

Optimal Algorithm

Determines a feasible scaling assignment σ such that the power consumption, , is minimized.

 

Chun-Han Lin, NTNU

Input :A region set R with an adjacency matrix A, as well as critical scaling ratios c( ) and A differential constant dOutput :A feasible assignment σσ ← c;Q ←R;While Q≠∅ do ← remove from Q a region whose key is maximum for all do if A[i,j] = 1 then σ[j] ← max(σ[j], )return σ

 

Slide41

Conversion Software

Image converter

Image editing software

Power-Saving Mode

OLED mobile device

Chun-Han Lin, NTNU

Slide42

Experiment Setup

4 images on Samsung Galaxy Tab 7.7Different characteristics in terms of luminance and saliencyPerformance MetricsExecution time (second) and power consumption (watt)ComparisonA grid-based approach revised based on that in a DAC’12 paper.

Chun-Han Lin, NTNU

Slide43

GRID vs. CURA

Execution time (seconds)Power consumption (watts)Visual quality See a video demo

GRIDCURAImage Converter27~2197.6~8.8Power-Saving Mode0.97~4.770.72~0.811

GRIDCURAImage Converter237~648284~572Power-Saving Mode362~797305~595

*

PSM uses Lanczos resampling to scale down the resolution for speedup at a cost of less power saving.

Chun-Han Lin, NTNU

Slide44

Outline

IntroductionLiquid Crystal DisplaysOrganic Light-Emitting Diode DisplaysConclusion

Chun-Han Lin, NTNU

Slide45

Conclusion

We raise the concept of cloud-based energy-saving services and have developed the dynamic backlight scaling service for mobile LCDsWith the service is applied, an HTC Desire mobile phone can save 18-31% backlight energy when browsing videos on YouTubeWe introduce visual attention into the quality-retaining power-saving design on mobile OLED displaysWe present CURA to realize the notion. Samsung Galaxy Tab 7.7 can save 38-42% OLED power while retaining visual quality

Chun-Han Lin, NTNU


About DocSlides
DocSlides allows users to easily upload and share presentations, PDF documents, and images.Share your documents with the world , watch,share and upload any time you want. How can you benefit from using DocSlides? DocSlides consists documents from individuals and organizations on topics ranging from technology and business to travel, health, and education. Find and search for what interests you, and learn from people and more. You can also download DocSlides to read or reference later.
Youtube