A 201.4 GOPS real-time multi-object recognitionprocessor is presented with a three-stage pipelined architecture.Visual perception based multi-object recognition algorithm isapplied to give multiple attentions to multiple objects in the inputimage. For human-like multi-object perception, a neural perceptionengine is proposed with biologically inspired neural networksand fuzzy logic circ
44circuits.Inhardwarearchitecture,athree-stagepipelinedar-chitecturehasbeenproposedtomaximizethethroughputofrecognitionprocessing.Thethreeobjectrecognitiontasksareexecutedinthepipelineandtheexecutiontimesofthethreetasksarebalancedforef cientpipeliningbasedonintelligentworkloadestimations.Inaddition,a118.4GB/smulti-castingnetwork-on-chiphasbeenproposedforcommunicationarchi-tecturewithincorporatingoverall21IPblocksoftheprocessor.Finally,workload-awaredynamicpowermanagementwasperformedforlow-powerobjectrecognition.The49mmchipcontains3.7Mgatesand396KBon-chipSRAMina
0.13mCMOSprocess.Withademonstrationsystem,thefabricatedchipachieves60frame/secmulti-objectrecognition
upto10differentobjectsforVGA
(640
480)videoinputwhiledissipating496mWat1.2V.Theobtained8.2mJ/frameenergydissipationis3.2timeslowerthanthestate-of-the-artrecognitionprocessor.
REFERENCES
[1]S.Kyoetal.,“A51.2GOPSscalablevideorecognitionprocessorfor
intelligentcruisecontrolbasedonalineararrayof1284-wayVLIWprocessingelements,”IEEEJ.Solid-StateCircuits,vol.38,no.11,pp.1992–2000,Nov.2003.
[2]A.Abboetal.,“XETAL-II:A107GOPS,600mWmassively-parallel
processorforvideosceneanalysis,”IEEEJ.Solid-StateCircuits,vol.43,no.1,pp.192–201,Jan.2008.
[3]D.Kimetal.,“An81.6GOPSobjectrecognitionprocessorbasedon
NoCandvisualimageprocessingmemory,”inProc.IEEECustomIn-tegratedCircuitsConf.(CICC),Apr.2007,pp.443–446.
[4]K.Kimetal.,“A125GOPS583mWnetwork-on-chipbasedparallel
processorwithbio-inspiredvisualattentionengine,”IEEEJ.Solid-StateCircuits,vol.44,no.1,pp.136–147,Jan.2009.
[5]J.-Y.Kimetal.,“A201.4GOPS496mWreal-timemulti-objectrecog-nitionprocessorwithbio-inspiredneuralperceptionengine,”inIEEEISSCCDig.Tech.Papers,Feb.2009,pp.150–151.
[6]D.G.Lowe,“Distinctiveimagefeaturesfromscale-invariantkey-points,”http://www.77cn.com.cnputerVision,vol.60,no.2,pp.91–110,Jan.2004.
[7]S.Leeetal.,“Thebrainmimickingvisualattentionengine:An802
60digitalcellularneuralnetworkforrapidglobalfeatureextraction,”inIEEESymp.VLSICircuitsDig.,Jun.2008,pp.26–27.
[8]L.Ittietal.,“Amodelofsaliency-basedvisualattentionforrapidscene
analysis,”IEEETrans.PatternAnal.MachineIntell.,vol.20,no.11,pp.1254–1259,Nov.1998.
[9]M.Kimetal.,“A22.8GOPS2.83mWneuro-fuzzyobjectdetection
engineforfastmulti-objectrecognition,”inIEEESymp.VLSICircuitsDig.,Jun.2009,pp.260–261.
[10]S.A.Nene,S.K.Nayar,andH.Murase,ColumbiaObjectImageLi-brary(Coil-100),ColumbiaUniversity,NewYork,TechnicalReportCUCS-006-96,Feb.1996.
[11]S.Agarwaletal.,“Learningtodetectobjectsinimagesviaasparse,
part-basedrepresentation,”IEEETrans.PatternAnal.MachineIntell.,vol.26,no.11,pp.1475–1490,Nov.2004.
[12]J.-Y.Kimetal.,“A66frame/sec38mWnearestneighbormatching
processorwithhierarchicalVQalgorithmforreal-timeobjectrecogni-tion,”inProc.IEEEA-SSCC,Nov.2008,pp.177–180.
[13]P.Pirsch,N.Demassieux,andW.Gehrke,“VLSIarchitectures
forvideocompression—Asurvey,”Proc.IEEE,vol.83,no.2,pp.220–246,Feb.1995.
[14]D.O.Hebb,TheOrganizationofBehavior.NewYork:Wiley,1949.[15]S.-J.Leeetal.,“An800MHzstar-connectedon-chipnetworkforap-plicationtosystemsonachip,”inIEEEISSCCDig.Tech.Papers,2003,pp.468–489.
[16]K.Leeetal.,“Low-powernetworks-on-chipforhigh-performanceSoC
design,”IEEETrans.VLSISyst.,vol.14,no.2,pp.148–160,Feb.2006.[17]K.Kimetal.,“A76.8GB/s46mWlow-latencynetwork-on-chipfor
real-timeobjectrecognitionprocessor,”inProc.IEEEA-SSCC,Nov.2009,pp.177–180.
[18]J.N.Seizovic,“Pipelinesynchronization,”inProc.IEEEASYNC,Nov.
1994,pp.87–96.
IEEEJOURNALOFSOLID-STATECIRCUITS,VOL.45,NO.1,JANUARY2010
[19]M.Keatingetal.,LowPowerMethodologyManualforSystemonChip
Design.NewYork:Springer,2007.
[20]B.Khailanyetal.,“Aprogrammable512GOPSstreamprocessorfor
signal,image,andvideoprocessing,”inIEEEISSCCDig.Tech.Pa-pers,Feb.2007,pp.
272–273.
Joo-YoungKim(S’05)receivedtheB.S.andM.S.degreesinelectricalengineeringandcomputersci-encefromtheKoreaAdvancedInstituteofScienceandTechnology(KAIST),Daejeon,Korea,in2005and2007,respectively,andiscurrentlyworkingto-wardthePh.D.degreeinelectricalengineeringandcomputerscienceatKAIST.
Since2006,hehasbeeninvolvedwiththedevelop-mentoftheparallelprocessorsforcomputervision.Currently,hisresearchinterestsareparallelarchitec-ture,sub-systems,andVLSIimplementationforbio-inspiredvision
processor.
MinsuKim(S’07)receivedtheB.S.andM.S.de-greesinelectricalengineeringandcomputersciencefromtheKoreaAdvancedInstituteofScienceandTechnology(KAIST),Daejeon,Korea,in2007and2009,respectively.HeiscurrentlyworkingtowardthePh.D.degreeinelectricalengineeringandcom-puterscienceatKAIST.
Hisresearchinterestsincludenetwork-on-chipbasedSoCdesignandbio-inspiredVLSIarchitectureforintelligentvision
processing.
SeungjinLee(S’06)receivedtheB.S.andM.S.de-greesinelectricalengineeringandcomputersciencefromtheKoreaAdvancedInstituteofScienceandTechnology(KAIST),Daejeon,Korea,in2006and2008,respectively.HeiscurrentlyworkingtowardthePh.D.degreeinelectricalengineeringandcom-putersciencefromKAIST.
Hispreviousresearchinterestsincludelow-powerdigitalsignalprocessorsfordigitalhearingaidsandbodyareacommunication.Currently,heisinvesti-gatingparallelarchitecturesforcomputervisionpro-
cessing.
JinwookOh(S’08)receivedtheB.Sdegreeinelec-tricalengineeringandcomputersciencefromSeoulNationalUniversity,Seoul,Korea,in2008.Heiscur-rentlyworkingtowardtheM.S.degreeinelectricalengineeringandcomputerscienceatKAIST,Dae-jeon,Korea.
Hisresearchinterestsincludelow-powerdigitalsignalprocessorsforcomputervision.Recently,heisinvolvedwiththeVLSIimplementationofneuralnetworksandfuzzy
logics.
KwanhoKim(S’04)receivedtheB.S.andM.SdegreesinelectricalengineeringandcomputersciencefromtheKoreaAdvancedInstituteofSci-enceandTechnology(KAIST)in2004and2006,respectively.HeiscurrentlyworkingtowardthePh.D.degreeinelectricalengineeringandcomputerscienceatKAIST.
In2004,hejoinedtheSemiconductorSystemLaboratory(SSL)atKAISTasaResearchAssistant.HisresearchinterestsincludeVLSIdesignforobjectrecognition,architectureandimplementationof
NoC-basedSoC.