版權(quán)說明:本文檔由用戶提供并上傳,收益歸屬內(nèi)容提供方,若內(nèi)容存在侵權(quán),請進(jìn)行舉報或認(rèn)領(lǐng)
文檔簡介
RochesterInstituteofTechnology
RITDigitalInstitutionalRepository
Theses
12-2021
Camera-baseddeeplearningAIassistantsystemforbasketballtraining
GuangkunZeng
gz4641@
Followthisandadditionalworksat:
/theses
RecommendedCitation
Zeng,Guangkun,"Camera-baseddeeplearningAIassistantsystemforbasketballtraining"(2021).Thesis.RochesterInstituteofTechnology.Accessedfrom
ThisThesisisbroughttoyouforfreeandopenaccessbytheRITLibraries.Formoreinformation,pleasecontact
repository@
.
CAMERA-BASEDDEEPLEARNINGAIASSISTANTSYSTEMFORBASKETBALLTRAINING
RIT
Camera-baseddeeplearningAIassistantsystemforbasketballtraining
By
GuangkunZeng
AThesisSubmittedinPartialFulfillmentoftheRequirementsfortheDegreeofMasterofFineArtinVisualCommunicationDesign
School/DepartmentofDesignCollegeofArtandDesignRochesterInstituteofTechnology
Rochester,NYDecember,2021
ThesisApproval
Camera-baseddeeplearningAIassistantsystemforbasketballtraining
ThesisTitle
GuangkunZeng
ThesisAuthor
Submittedinpartialfulfillmentoftherequirementsforthe
degreeofMasterofFineArts
TheSchoolofDesign|VisualCommunicationDesign
RochesterInstituteofTechnology|Rochester,NewYork
MikeStrobert ChiefThesisAdvisor
Name Title
ElectronicSignature:UseAdobeAcrobat
AssociateThesisAdvisor
Name Title
ElectronicSignature:UseAdobeAcrobat
Name(Optionalforsomeprograms)
selectaTitle
Title(Optionalforsomeprograms)
ElectronicSignature:UseAdobeAcrobat(Optionalforsomeprograms)
selectaTitle
Name(Optionalforsomeprograms) Title(Optionalforsomeprograms)
ElectronicSignature:UseAdobeAcrobat(Optionalforsomeprograms)
CAMERA-BASEDDEEPLEARNINGAIASSISTANTSYSTEMFORBASKETBALLTRAINING
Abstract
TheYOLO,aComputerVisionAlgorithms,isbroughtouttoanalyzethebasketballplayer’sstatusasadataset.Itcanrecordtheplayers’behavioronthecourtincludingdribbling,shooting,andrunning.Inthisway,theappcouldcollectthefieldgoalyoumadeandmissed.First,youshouldusethisapptorecordavideoofyourshoottraining.Afterthat,theAIwouldanalyzeandbringsouta3dvirtualdiagraminterpretyourperformance.Thisdiagramwillshowthehotzoneandcoldzoneforyourfieldgoal.Also,thetrackofyourballwillbedisplayedonthevideosothatyoucanknowiftheangleofyourshootingistoolowortoohigh.Intheend,theAI-basedonmachinelearningwillgiveoutaplanaccordingtoyourperformanceonshooting.
Asatrainingmobileapplicationsupportedbycamera-basedactionrecognition,thetargetaudienceisthebasketballamateurplayerswhodon’thavetheresourcesasproplayersdo.Thisprojectwillbedesignedasanewtrainingexperienceandwillbedeliveredasapromovideothatshowshowtousetheapplicationandalsothescenariopeopleuse.
Keywords
MR,virtualenvironment,camera-based,trainingsystem,deep-learning,machinelearning,motioncapture,AI
CriticalAnalysisandSummary
Context
Manybasketballenthusiastswanttoupgradetheirskillstoafarbetterlevel,butunfortunately,noteveryonehasthefundstohireprofessionalcoachestotrainonthebestcourt.Andwhat’smore,manyamateurplayersdonotevenknowhowtopracticeinaproperwaybecausethevideotutorialisnotasintuitiveastherealcoachorastheprofessionalhumancoachis.What’smore,itisalmostimpossibleforthebasketballplayertoquantifyhowgoodhis/hershootingis.Butpeopledoneedverygoodshootingformandtechniquetoshootconsistently.Howmightweimproveplayers’trainingexperience?
Nowadays,MR(mixedreality)haschangedourlife.ResearchbyVisualCapitalistprojectsthattheXRmarketwillbeworth$209billionby2022,markinganeight-foldincreasefrom20181.Furthermore,63%ofshareholdersinXRtechnologycompaniesbelievethetechnologywillbemainstreamby2024.Atthesametime,MixedRealityisquitewithinthemiddlebutalsothelongertermoftheentertainmentindustry.2Thistypeoftechnologycouldprovideacustomizedanduniquetrainingexperienceevencomparedtothebestcourtandtrainerscoulddo.3
TheMixedRealityisthetechnologythatcouldprovideanimmersiveexperienceanddirectdatavisualization.IproposetobringoutaMixedRealitymobileapplicationthatcangenerate3dvirtualdiagramofthebasketballcourttoassistplayerstounderstandtheirshootingperformance.Atthesametime,theappwillcollecttheplayer’sdataandmakethetrainingplanaccordingtotheAI-basedanalysissystem.
Methodology
Toimplementthissystem,multipletechnologieswillbeintegrated.TheYOLO,aComputerVisionAlgorithmswillrecognizetheballtraceandcollectdataofyourtraining.4Themotioncapturewith3Dvirtualenvironmenttechnologywillinterprettheshootinggestureandmovement.Thenthemachinelearningwillgenerateaspecifictrainingplan.
Thephaseofinterviewanduserresearchwillnotproceedbecausethisdesignputsmoreemphasisonconceptualdesign.Thisdesignfocuseson3problemsthatplayersoftenmeet.First,it’shardforamateurplayerstoquantifytheirskilllevel.Second,noteveryplayercanmakeapreciseplanasprofessional
1Emrich,T.(2020,February25).20for2020:AugmentedRealityTrendsandHowTheyMayPlayOutThisYear[Weblogpost].Retrievedfrom
/@tomemrich/20-augmented-reality-trends-to-keep-
an-eye-on-in-2020-d2b0258edbb
2Terry,Q.(2019,July23).ARiselevatingtheplayingfieldforsportsbycreatingsmartertrainingmethods.Retrievedfrom/futuresin/ar-is-elevating-the-playing-field-for-sports-by-creating-smarter-training-methods-77db01a84d64
3Lee,David."OurFirstShot(s)."Medium(blog).July17,2018./nex-team/our-first-shot-s-272c67d0349d.
4Terry,Q.(2019,July23).ARiselevatingtheplayingfieldforsportsbycreatingsmartertrainingmethods.Retrievedfrom/futuresin/ar-is-elevating-the-playing-field-for-sports-by-creating-smarter-training-methods-77db01a84d64
coachescando.Third,playersarenotsurethepracticeisexecutedperfectly.Peoplealsosometimesforgothowmanyshotsthey’vemade.
Threesolutionshavebeengivenout:First,A3ddiagramofyourplayscanhelpuserstoquantifytheirperformance.ThenAIDeeplearningcangiveyouadviceasgoodasacoachdoesorevenbetter.Finally,Asmartphonecamera-basedreal-timesystemcanrecordyourplayswithnomistakes.
Astheresultoftheproblemsolutions,HoopLabidentified3specificdesigngoalsthatareusedtoprototype:
Friendlytouse
Createanexperiencethatuserscanenjoy.Userscancustomizetheir3davatarandpickupfavoriteclothandhaircutsforit.Theavatarwillberiggedandmotioncapturedbyusersasawaytounderstandyourshootingmovement.
Easytounderstand
A360degrees3Ddiagramwillbegeneratedaccordingtotheplayrecording.Itisabletoturnyourstillcamerarecordinginto3dversionenvironment.TheAIwillcalculatethedistancebetweenobjects(player-player,shooter-hoop,defender-ball)
Convenienttorecord
Theuseronlyneedsasmartphonecameratodothebasicfunction.Usingasmartwatchtounlockadvancedfeatures.
Afterthreeroundsofprototyping,thecorrespondinginterfacelo-fiwireframesaredesigned.Basedonthoseinteractionwireframes,theoutputoftheUIvisualwireframeisfinallycompleted.ApromovideothatcombinedUIelementsandapplicationfunctionsisalsobemade.
Agamedesigntheoryisalsobroughtinforthissystem.Userscanearncoinsbyfinishingthetasks.Thentheycanusegamecoinstounlockavatarsandnewadvancedtasks.Thecirculareconomyisformedinthisprocess.
Ithinktherewillbesustainableiterationstoevaluatemyproposal.Atphaseone,whichistheinitialstateofaproduct,alargenumberofA/Btestswillbegiventotheusers.AccordingtotheA/Btestresultsthatwecollect,itcanbedecidedwhichfeatureisbetterforusers.Inphasetwo,theproductstartstooperatesmoothly,thenplentyofdatacanbeanalyzedbymachinelearning.DuetoAI,wecandecidewhatnewfeaturesthatmaysatisfyusersshouldbebuiltfurther.
Conclusion
It’sdifficulttoquantifytheexperienceofthesystem.Ifthesystemcouldpresentacustomizedtrainingplanfortheuserandcreateavirtual3denvironmentofarealcourtthattheplayerdoesn’tfeelstrangeaboutit,thisdesignwillbeasatisfiedsolution.
However,withthelimitofreal-timerendertechnologyandhardwarehashingpower,itisstillimpossibletoimplementanapplicationlikethisinamobilephone.ThisprojectisanapproachthatrepresentsonepossibilityofwhatAIalgorithmscandoforsportsgames.
AppendixA:ExpandedThesisDefensePresentation
Initialdesign
Designintension
溫馨提示
- 1. 本站所有資源如無特殊說明,都需要本地電腦安裝OFFICE2007和PDF閱讀器。圖紙軟件為CAD,CAXA,PROE,UG,SolidWorks等.壓縮文件請下載最新的WinRAR軟件解壓。
- 2. 本站的文檔不包含任何第三方提供的附件圖紙等,如果需要附件,請聯(lián)系上傳者。文件的所有權(quán)益歸上傳用戶所有。
- 3. 本站RAR壓縮包中若帶圖紙,網(wǎng)頁內(nèi)容里面會有圖紙預(yù)覽,若沒有圖紙預(yù)覽就沒有圖紙。
- 4. 未經(jīng)權(quán)益所有人同意不得將文件中的內(nèi)容挪作商業(yè)或盈利用途。
- 5. 人人文庫網(wǎng)僅提供信息存儲空間,僅對用戶上傳內(nèi)容的表現(xiàn)方式做保護(hù)處理,對用戶上傳分享的文檔內(nèi)容本身不做任何修改或編輯,并不能對任何下載內(nèi)容負(fù)責(zé)。
- 6. 下載文件中如有侵權(quán)或不適當(dāng)內(nèi)容,請與我們聯(lián)系,我們立即糾正。
- 7. 本站不保證下載資源的準(zhǔn)確性、安全性和完整性, 同時也不承擔(dān)用戶因使用這些下載資源對自己和他人造成任何形式的傷害或損失。
最新文檔
- 二零二五年度歷史學(xué)論文版權(quán)一次性轉(zhuǎn)讓及后續(xù)研究資助合同3篇
- 二零二五年度科研機(jī)構(gòu)兼職教師任聘合同2篇
- 2025年度農(nóng)村出租屋租賃與農(nóng)村文化產(chǎn)業(yè)發(fā)展合作合同
- 二零二五年度高校畢業(yè)生實(shí)習(xí)就業(yè)三方服務(wù)合同2篇
- 二零二五年度住宅小區(qū)電梯加裝工程驗(yàn)收合同3篇
- 二零二五年度委托監(jiān)護(hù)與協(xié)議監(jiān)護(hù)教育成長規(guī)劃全面服務(wù)合同3篇
- 全新清算協(xié)議合同-二零二五年度清算解決方案3篇
- 二零二五年度房地產(chǎn)開發(fā)商收購合作伙伴合同3篇
- 2025年度農(nóng)村手房買賣合同模板(含土地抵押權(quán))
- 2025年度農(nóng)村房屋置換與農(nóng)村產(chǎn)業(yè)發(fā)展扶持合同
- 建設(shè)工程見證取樣管理規(guī)范
- 中國成人血脂異常防治指南解讀
- 醫(yī)學(xué)專家談靈芝孢子粉課件
- 彈性力學(xué)19年 吳家龍版學(xué)習(xí)通超星課后章節(jié)答案期末考試題庫2023年
- 有沒有租學(xué)位的協(xié)議書
- 車載智能計算芯片白皮書
- 住宅小區(qū)綠化管理規(guī)定
- 土建工程定額計價之建筑工程定額
- 2022年7月云南省普通高中學(xué)業(yè)水平考試物理含答案
- 學(xué)校安全工作匯報PPT
- 一年級語文上冊《兩件寶》教案1
評論
0/150
提交評論