Spatially-Attentive Patch-Hierarchical Network for Adaptive Motion
Spatially-Attentive Patch-Hierarchical Network for Adaptive Motion Deblurring. Maitreya Suin?. Kuldeep Purohit?. A. N. Rajagopalan.
Densely Connected Hierarchical Network for Image Denoising
Our proposed network improves the image denoising performance by applying the hierarchical architecture of the modified U-Net; this enables our network to use a
Hierarchical organization in complex networks - Erzsébet Ravasz
14-Feb-2003 In hierarchical networks the degree of clustering characterizing the different groups follows a strict scaling law
Objectives Converged Networks Hierarchical Network Design
Match the appropriate Cisco switch to each layer in the hierarchical network design model. Converged Networks. ? Combining voice and video communications on a
Network Intrusion Detection: Based on Deep Hierarchical Network
03-Apr-2019 By designing a reasonable network cascading method we can train our proposed hierarchical network at the same time instead of training two.
Network Intrusion Detection Combined Hybrid Sampling With Deep
24-Feb-2020 INDEX TERMS Network intrusion detection hybrid sampling
Fast Deep Multi-Patch Hierarchical Network for Nonhomogeneous
a fast Deep Multi-patch Hierarchical Network to restore. Non-homogeneous hazed images by aggregating features from multiple image patches from different
RTHN: A RNN-Transformer Hierarchical Network for Emotion Cause
Transformer Hierarchical Network (RTHN) to en- code and classify multiple clauses synchronously. RTHN is composed of a lower word-level encoder.
A Deep Hierarchical Network for Packet-Level Malicious Traffic
17-Nov-2020 INDEX TERMS Network intrusion detection deep learning
Hierarchical Network for Facial Palsy Detection
2. We propose the Hierarchical Detection Network (HDN) which consists of three component networks
[PDF] Chapter 1: Hierarchical Network Design
Describe how a hierarchical network model is used to design networks ? Explain the structured engineering principles for network design:
[PDF] Hierarchical Network Design - Pearsoncmgcom
13 mar 2014 · This topic discusses the three functional layers of the hierarchical network model: the access distribution and core layers Network Hierarchy
[PDF] Hierarchical Network Design - DTU Informatics
The design of hierarchical networks involves clustering of nodes hub selection and network design i e selection of links and routing of flows Hierarchical
[PDF] AN ALGORITHM FOR HIERARCHICAL NETWORK DESIGN
We consider a centralized concentrator-based network (Gavish 1992) in a 2-level hierarchical configuration A set of terminal nodes is served by a set of
[PDF] The Switch Hierarchical Network Design Model (SHiNDiM)
Building up a strong hierarchical network requires installing switches in the correct position in the network such as the access distribution and core layers
[PDF] Small Enterprise Design Profile (SEDP)—Network Foundation Design
This design employs the four key design principles of hierarchy modularity resiliency and flexibility Figure 1 Three-Tier Hierarchical Model Each layer in
A hierarchical network model for network topology design using
7 mar 2023 · PDF Network topology design has directly impact on network construction costs and network performance Majority of current network
An Adaptive Hierarchical Network Model for Studying the Structure
9 mar 2023 · PDF The interdependence of financial institutions is primarily responsible for creating a systemic hierarchy in the industry
[PDF] The Hierarchical Network Topology Management System based on
The Hierarchical Network Topology Management System based on Managed Object and View Mechanism Hui-Qin Jin Man-Gui Liang*
MaitreyaSuin
?KuldeepPurohit?A.N.RajagopalanIndianInstituteof TechnologyMadras, India
Abstract
Thispapertac klesthepr oblemofmotiondeblurringof
dynamicscenes.Although end-to-endfullycon volutional designshaver ecentlyadvancedthe state-of-the-artinnon- uniformmotiondeblurri ng, theirperformance-complexity trade-offisstillsub-optimal.Existingapproac hesachie ve alarg ereceptivefieldbyincreasingthenumber ofgeneric convolutionlayersandk ernel-size,butthiscomes atthee x- penseofof theincrease inmodelsize andinference speed. Inthiswork, wepr oposeanef ficientpixeladaptive and featureattentivedesignforhandling larg eblurvariations acrossdifferentspatial locationsandprocesseachtestim- ageadaptively.Wealsopr oposeaneffectivecontent-aware global-localfiltering modulethatsignificantlyimproves performancebyconsidering notonlyglobal dependencies butalsobydynamicallye xploitingneighboringpixel infor- mation.We useapatch-hierarc hicalattentivear chitecture composedofthe abovemodule thatimplicitlydisco versthe spatialvariationsin theblurpr esentinthe inputimag eand inturn,performs localandglobal modulationofinterme- diatefeatures. Extensivequalitativeandquantitativecom- parisonswithprior artondeblurring benchmarksdemon- stratethatourdesignof ferssignificant improvements over thestate-of-the-artin accuracyas wellasspeed.1.Introduction
Motion-blurredimagesform duetorelati vemotion dur- ingsensore xposureandare favoredbyphotographers and artistsinman ycasesfor aestheticpurpose,butseldomby computervisionresearchers, asmany standardvisiontools includingdetectors,track ers,andfeature extractorsstruggle todealwith blur.Blind motiondeblurringis anill-posed problemthataims torecov erasharp imagefroma given imagedegraded duetomotion-inducedsmearingofte x- tureandhigh-frequenc ydetails. Duetoitsdiverse applica- tionsinsurv eillance,remotesensing, andcamerasmounted ?Equalcontribution.10-1100101
Runtime for an HD image (seconds)
28.52929.53030.53131.532
PSNR (dB)
Nah CVPR17
Kupyn CVPR18
Tao CVPR18
Ours(a)
Zhang CVPR18
Zhang CVPR19
Gao CVPR19
Kypyn ICCV2019
Ours(b)
Figure1.Comparison ofdifferent methodsinterms ofaccuracy andinferencetime. Ourapproachoutperforms allprevious meth- ods. onhand-heldand vehicle-mountedcameras, deblurringhas gatheredsubstantialattentionfromcomputer visionandim- ageprocessingcommunities inthepast twodecades. Majorityoftraditional deblurringapproachesare based onvariational model,whosekeycomponent isthere gular- izationterm.The restorationqualitydepends ontheselec- etersinv olvinghighlynon-convexoptimizationsetups[14]. achallengingcomputer visionproblemas blursarisefrom varioussourcesincludingmoving objects,camerashak e anddepthv ariations,causingdif ferentpixelstocapturedif- whilegeneralizingacross differenttypes ofreal-world ex- amples,whereblur isfar morecomplex thanmodeled[3].Recentworks basedondeepconvolutional neuralnet-
works(CNN)have studiedthebenefits ofreplacingthe imageformationmodel withaparametric modelthatcan betrainedto emulatethenon-linear relationshipbetween blurred-sharpimagepairs .Suchw orks[13]directlyregress todeblurredimage intensitiesando vercomethe limited representativecapabilityofvariationalmethodsin describ- ingdynamicscenes. Thesemethodscan handlecombined effectsofcameramotionand dynamicobjectmotion and 1 3606achievestate-of-the-artresultsonsingleimagedeblurring task.They havereacheda respectablereductioninmodelsize,but stilllackinaccuracyand arenotreal-time.
ExistingCNN-basedmethods have twomajor limita-
tions:a)W eightsofthe CNNarefixedandspatially in- variantwhichmaynotbe optimalfordif ferentpixels ina dynamicallyblurredscene (e.g.,sky vs.moving carpix- els).Thisissue isgenerallytackled bylearninga highly non-linearmappingby stackingalar genumberof filters. Butthisdrastically increasesthecomputational costand memoryconsumption.b) Ageometricallyuniform recep- tivefieldissub-optimalforthetask ofdeblurring.Lar ge imageregions tendtobeusedtoincrease thereceptiv efield eventhoughtheblurissmall.This inevitablyleads toa networkwithalarge numberoflayers andahigh compu- tationfootprintwhich slowsdo wnthecon vergenceof the network. tivefieldandtheaccuracyof anetwork isanon-tri vialtask (seeFig.1). Ourwork focusesonthe designofef ficientand interpretablefilteringmodules thatoffer abetteraccurac y- speedtrade-off ascomparedtosimplecascadeof convolu- tionallayers.W einv estigatemotion-dependentadaptability withinaCNN todirectlyaddress thechallengesi nsingle imagedeblurring.Since motionbluris inherentlydirec- tionalanddif ferentforeach imageinstance,adeblurring networkcanbenefitfromadapting totheblur presentin eachinputtest image.We deploycontent-a waremodules whichadjustthe filtertobe appliedandthe receptive field ateachpix el.Ouranalysis showsthatthe benefitsofthese dynamicmodules forthedeblurringtaskaretw o-fold:i) Cascadeofsuch layersprovides alarge anddynamically adaptivereceptivefield.Directionalnature ofblurrequires adirectionalrecepti vefield, whichanormalCNNcannot achievewithinasmallnumberoflayers. ii)Itef ficiently enablesspatiallyv aryingrestoration,since changesinfilters andfeaturesoccur accordingtothe blurinthe localregion. Noprevious workhasinv estigatedincorporatingaw areness ofblur-v ariationwithinanend-to-endsingleimagedeblur- ringmodel.Followingthestateoftheartin deblurring,weadopt a
multi-patchhierarchicaldesign todirectlyestimate there- storedsharpimage. Insteadofcascading alongthedepth, weintroducecontent-a warefeature andfiltertransforma- tioncapability throughaglobal-localattentive moduleand residualattentionacross layerstoimpro veperformance. Thesemoduleslearn toexploit thesimilarityin themotion betweendifferent pixelswithinanimageand arealsosen- sitivetoposition-specificlocalcontext.Theefficienc yofourarchitectureisdemonstrated
throughacomprehensi vee valuationontwobenchmarks andcomparisonswith thestate-of-the-art deblurringap- proaches.Ourmodel achieves superiorperformancewhile Figure2.Ov erallarchitectureof ourproposednetwork.CAblock representscrossattention betweendiff erentlev elsofencoder - decoderanddif ferentlev els.Alltheresblockcontains onecon- tentaw areprocessingmodule.Symbol"+"denoteselementwise summation. beingcomputationallymore efficient.The majorcontribu- tionsofthis workare:Weproposean efficientdeblurringdesignbuiltonnew convolutionalmodulesthatlearnthetransformationof featuresusingglobal attentionandadapti velocal fil-ters.We showthatthesetwo branchescomplementeachotherand resultinsuperior deblurringperfor- mance.Moreov er,theefficientdesignofattention-moduleenablesus touseit throughoutthenetw orkwithouttheneed forexplicit downsampling.
Wefurtherdemonstratetheef ficacyof learningcross-attentionbetweenencode-decoder aswellas differentlevelsinourdesign.
Weprovideextensi veanalysisandev aluationsondy-namicscenedeblurring benchmarks,demonstratingthatourapproach yieldsstate-of-the-artresults while
being3×fasterthanthenearestcompetitor [26].2.Proposed Architecture
Todate,thedriving forcebehindperformance improve-
ersandlar gerfilterswhich assistinincreasingthe"static" receptivefieldandthegeneralizationcapabilityof aCNN. networkperformancedoesnotal waysscale withnetwork depth,asthe effectiv ereceptiv efieldofdeepCNNsismuch smallerthanthe theoreticalvalue (investig atedin[12]). Weclaimthatasuperior alternative isadynamic frame- workwhereinthefilteringand thereceptiv efieldchange 3607acrossspatiallocations andalsoacross differentinput im-ages.Oure xperimentsshow thatthisapproachisacon- siderablybetterchoice duetoits task-specificefficac yandutilityforcomputationally limitedenvironments. Itdeliv ersconsistentperformanceacross diverse magnitudesofblur .
Althoughprevious multi-scaleandscale-recurrentmeth- odshav eshowngoodperformanceinremovingnon- uniformblur, theysufferfrome xpensiveinferencetime andperformancebottleneck whilesimplyincreasing model depth.Instead,inspired by[26], weadoptmulti-patch hi- erarchicalstructureas ourbase-model,which comparedto multi-scaleapproachhas theaddedadv antageofresidual- likearchitecturethatleadsto efficientlearning andfaster processingspeed.The overall architectureofour proposed networkisshownin Fig.2.W edividethenetwork into3 levelsinsteadof4asdescribedin [26].We foundthatthe relativeperformancegainduetothe inclusionofle vel4is negligiblecomparedtotheincrease ininferencetime and numberofparameters. Atthebottom level inputslicedinto4non-ov erlappingpatchesforprocessing,andaswegrad-
uallymov etowardshigherlevels,the numberofpatches decreaseandlo werlev elfeaturesareadaptively fusedusing attentionmoduleas shownin Fig.2.The outputoflevel1 isthefinal deblurredimage.Note thatunlike [26],wealso avoidcascadingofournetworkalong depth,asthat addsse- verecomputationalburden.Instead, weadvocate theuseof performanceimprov ementsovereventhedeepest stacked versionsoforiginalDMPHN[26]. Majorchangesincorpo- ratedinour designaredescribed next.Eachlev elofournetworkconsistsofanencoder anda
decoder.Boththeencoderand thedecoderaremadeofstan- dardconv olutionallayerandresidualblockswhereeachof theseresidualblocks contains1con volutionlayer followed byacontent-a wareprocessing moduleandanotherconvo- lutionallayer. Thecontent-awareprocessingmodule com- prisestwo branchesforglobalandlocalle velfeature pro- cessingwhichare dynamicallyfused attheend. Theresid- ualblocksof decoderandencoder areidenticale xceptfor theuseof crossattentionin decode r.W ehav ealsodesigned cross-levelattentionforeffective propagationof lowerle vel featuresthroughoutthe network.W ebegin withdescrib- ingcontent-aw areprocessingmodule,thenproceedtowards thedetaileddescription ofthetw obranchesand finallyhow thesebranchesare adaptively fusedatthe end.3.Content-Awar eProcessingModule
Incontrastto high-level problemssuchas classification anddetection[22], whichcanobtain largerecepti vefield bysuccessiv elydown-samplingthefeaturemapwithpool- ingorstrided convolution, restorationtaskslik edeblurring needfinerpix eldetailsthat cannotbeachieved fromhighlydownsampledfeatures.Mostofthe previousdeblurring ap-proachesusesstandard convolutional layersforlocal fil-teringandstack thoselayerst ogethertoincrease there-ceptivefield.[1]usesself-attentionandstandard convo- lutiononparallel branchandsho wsthatbest resultsareobtainedwhenboth featuresarecombined togethercom-paredtousing eachfeatureseparately .Inspiredby thisapproach,wedesign acontent-aw areglobal-local"pro- cessingmodulewhich dependingonthe input,deploys twoparallelbranchesto fuseglobaland localfeatures.The global"branchis madeofattention module.For decoder,thisincludesboth selfandcross-encoder -decoderattentionwhereasforencoder onlyself-attentionis used.For localbranchwedesign apixel-dependent filteringmodulewhich determinestheweight andthelocal neighbourhoodtoap- plythefilter adaptively .We describeindetailthesetwobranchesandtheir adaptive fusionstrategy inthefollowing
sections.3.1.Attention [21]innatural languageprocessingdomain, ithasbeen in- troducedinimage processingtasksas well[15,11]. The mainbuilding blockofthisarchitecturei sself-attention whichasthe namesuggestscalculates theresponseat apo- sitionina sequencebyattending toallpositions withinthe samesequence.Gi venan inputtensorofshape(C,H,W) itisflattened toamatrix z?RHW×Candprojectedto d aanddcdimensionalspacesusing embeddingmatrices WA,B?RHW×daandC?RHW×dcareknown asquery,
keyandvalue,respectively .Theoutput oftheself-attention mechanismfora singleheadcan beexpressed asO=softmax?ABT
⎷da? C(1)Themaindra wbackofthis approachisveryhighmemory
requirementdueto thematrixmultiplication ABTwhich requiresstoring ahighdimensionalmatrixofdimension (HW,HW)forimagedomain. Thisrequiresa largedo wn- samplingoperationbefore applyingattention.[15] and[17] usealocal memoryblockinstead ofglobalall-to-all for makingitpractically usable.[1]uses attentiononlyfrom thelayerwith thesmallestspatial dimensionuntilit hits memoryconstraints.Also, theseworks typicallyresorttoquotesdbs_dbs21.pdfusesText_27[PDF] hierarchical regression table apa
[PDF] hierarchical structure journal article
[PDF] hierarchy java example
[PDF] hierarchy of law reports
[PDF] hifly a321
[PDF] hifly a380 interior
[PDF] hifly a380 model
[PDF] high appellate court definition
[PDF] high court
[PDF] high efficiency boiler
[PDF] high level french adjectives
[PDF] high net worth individuals survey
[PDF] high paid jobs in demand uk
[PDF] high paying jobs in high demand uk