Web-Based Scientific Exploration and Analysis of 3D Scanned Cuneiform Datasets for ...

16 Pages·2017·8.47 MB·English

informa t ics
Article
Web-Based Scientific Exploration and Analysis
of 3D Scanned Cuneiform Datasets for
Collaborative Research
DenisFisseler1,∗,GerfridG.W.Müller2andFrankWeichert1
1 DepartmentofComputerScienceVII,TUDortmundUniversity,44227Dortmund,Germany;
[email protected]
2 DepartmentofAncientCultures,AncientNearEasternStudies,UniversityofWürzburg,
97070Würzburg,Germany;[email protected]
* Correspondence:denis.fi[email protected]
AcademicEditor:RobertoTherónSánchez
Received:8November2017;Accepted:9December2017;Published:12December2017
Abstract: Thethree-dimensionalcuneiformscriptisoneoftheoldestknownwritingsystemsanda
centralobjectofresearchinAncientNearEasternStudiesandHittitology. Animportantsteptowards
theunderstandingofthecuneiformscriptistheprovisionofopportunitiesandtoolsforjointanalysis.
This paper presents an approach that contributes to this challenge: a collaborative compatible
web-basedscientificexplorationandanalysisof3Dscannedcuneiformfragments. TheWebGL-based
conceptincorporatesmethodsforcompressedweb-basedcontentdeliveryoflarge3Ddatasetsand
highqualityvisualization. Tomaximizeaccessibilityandtopromoteacceptanceof3Dtechniques
inthefieldofHittitology,theintroducedconceptisintegratedintotheHethitologie-PortalMainz,
anestablishedleadingonlineresearchresourceinthefieldofHittitology,whichuntilnowexclusively
included2Dcontent. Thepapershowsthatincreasingtheavailabilityof3Dscannedarchaeological
datathroughaweb-basedinterfacecanprovidesignificantscientificvaluewhileatthesametime
findingatrade-offbetweencopyrightinducedrestrictionsandscientificusability.
Keywords: cuneiform;3Dviewer;WebGL;Hittitology;3Dcollation
1. Introduction
Recent advances in 3D scanning technology have led to increased efforts to create digital
reproductionsofarchaeologicalartifactssuchascuneiformtabletfragments,asshowninFigure1,
resulting in large databases of high-resolution 3D meshes [1]. However, the existence of these 3D
data-repositoriesdoesnotresulttothesameextentinanincreasedavailabilityofthedataforassociated
researchpurposes. Therefore,accessibilitytothescientificcommunityisanongoingissuewithlarge
scale archaeological data-repositories. The reasons for this accessibility gap range from legal and
political issues, tied to the copyright on the data and data preparation times/costs, to technical
challenges regarding visualization and transfer of large amounts of data [2]. As will be detailed
lateron,anincreasingamountofweb-based3Dlibrariesandefficient3Dcontenttransfertechniques
becomesavailable,whichleadstoanincreasingamountof3Ddigitalheritagecontentpresentedin
theweb. Thenon-research-orientedpresentationobjectivesinmanyoftheseprojectsresultinahigh
suitabilityformuseumpresentationbutatthesametimeinaverylimitedsuitabilityforscientific
metrologicalanalysis. Occurringrestrictionsregardingthescientificusemayoriginatefromfactors
suchaslowscanortextureresolution,missingmeasurementandlightingtools,restrictedviewport
navigation, insufficient shading capabilities for enhancing subtle features on untextured datasets
andsoftware/datalicensingincompatibilities. Astherequirementstothedataanditspresentation
Informatics 2017,4,44;doi:10.3390/informatics4040044 www.mdpi.com/journal/informatics
Informatics 2017,4,44 2of16
varydependingonthefieldofresearch,Section2willintroducetheusecaseofcuneiformfragment
collation with a specific set of traditional research methods and requirements to be targeted in a
scientifically oriented 3D cuneiform framework. This set of requirements will then be used in the
courseofSection3toconstructasuitable3Dframeworkforweb-basedanalysisofcuneiformscript.
TheaspectstobecoveredincludethechoiceofaWebGL(WebGraphicsLibrary)environment,analysis
of3Dcuneiformdatacharacteristicsandpreparation,visualizationandanexemplaryintegrationinto
anestablishedonlineresearcharchitectureinSection4. ThesubsequentevaluationinSection5shows
that,fortheapplicationscenarioofweb-based3Dcuneiformscriptanalysis,atrade-offbetweenlegal
andcopyrightaspectsononesideandfreeavailabilityandscientificusabilityontheothersidecan
befoundtogenerateaddedscientificvalue. Thepapercloseswithashortconclusionandanoutlook
regardingfurtherresearchinSection6.
(a) (b) (c)
Figure1. Examplesofcommoncuneiformrepresentations. Anautographyofcuneiformfragment
262/pisshownin(a),withacorrespondingphotographicdocumentationin(b)andavisualizationofa
3Dscanin(c).Source:[3].
2. CuneiformFragmentCollation
Cuneiformscriptis,asidefromEgyptianhieroglyphs,oneoftheoldestknownwritingsystems[2].
Itwasusedforoverthreemillenniastartingwithitsinventionaround3500BCEinAncientNearEast
byavarietyofculturesliketheSumerians,theBabylonians,theAkkadians,theElamites,theAssyrians
and the Hittites. In contrast to modern scripts, cuneiform manuscripts were usually created by
imprinting an angular stylus into tablets of moist clay resulting in small and often overlapping
tetrahedron shaped grooves, which makes cuneiform a three-dimensional script. Most known
remainsofcuneiformmanuscriptshavebeenpreservedinformof(inmanycasesfragmented)clay
tablets,withaquantityofover500,000discoveredartifacts,storedmostlyinmuseumcollections[1].
Asidefrommuseumpresentation,thesecuneiformfragmentsaresubjecttoscientificresearchinthe
fieldofAncientNearEasternStudies. Basedonthedeciphermentofcuneiformscript,thefragments
areanalyzedandtranslatedinordertoreassemblethemanuscriptsandextendourknowledgeofthis
partofhumanhistory.
Inphilology,includingAncientNearEasternStudiesofcuneiformfragments,collationdenotes
theverificationofatextagainstanoriginalasacentralresearchmethod. Thisincludes,amongstother
things,theverificationoftranscriptionsandtransliterations. Ideally,“original”referstotheoriginal
manuscript in the form of a real cuneiform tablet or fragment. However, accessibility issues and
artifactcorrosionoftenresultinthereplacementoftheoriginalartifactbyhandcopies,photographic
reproductions and, recently, 3D scanned digital reproductions. Some examples for these types of
reproductionsareshowninFigure1. Thefollowingwilldetailsomeessentialcollation-relatedaspects
tobeconsideredduringthedevelopmentofsuitableweb-basedaccessmethods.
Informatics 2017,4,44 3of16
2.1. ApplicationScenarios
Despite very individualistic philological working methods and work flows, some typical
applicationscenariosforcollationtaskscanbeidentified. Oneofthesescenariosistheinspection
of transliterations for missing signs and text lines, which frequently occur, especially on larger or
difficulttoreadmanuscripts. Thisincludescheckingwhetherthesignscanbereadasspecifiedinthe
transliteration. Anotherreasonforconductingacollationarewell-foundeddoubtsaboutthecorrect
readingoftextpassagesorsigns. Inthiscontext,collationcanoperateonamoresemanticlevelto
determinehowthesigns,consistingofidentifiedwedges,mustbereador,onamoregeometry-oriented
level, to determine what wedges can be found on the artifact and which signs can be derived.
The second case is particularly important regarding damaged text passages and gaps. Resolving
thesekindsofgeometricissuesincludescheckingwhetherthecomplementedsignshavethecorrect
sizewithrespecttocomparablesignsonthesametablet. Additionally,theremaininggeometricclues
arecheckedagainstthegeometryofcomplementedsignsandpossiblesignalternatives,whichare
thenvalidatedregardingthetextualandpaleographiccontext.
2.2. CollationofVariousMediaTypes
Forcollationpurposes,arealartifactisusuallyviewedfromdifferentanglestoestimatewedge
depthsandsurfacedistancesofdamagedareas. Thishelpswithdistinguishingscratchesandfracture
facesfromremainsofcuneiformwedges. Thistechniqueisinmostcasescomplementedbyusinga
lightsourcetoenhancetheperceptibilityofgeometricfeaturesbyshadowcasting. Asidefromthe
depthofgrooves,thesurfacefinishcanhelpwiththedifferentiationofscratchesandwedges. Itcanbe
determinedtoalimitedextendbyemployingamagnifyingglass. Anadditionalfrequentissueinthis
regardisthepresenceofnon-uniformcoloredsurfacetexturesontheartifacts,whichcanexertahuge
negativeeffectonthevisualperceptibilityofgeometricfeatures.
Capability-wise,collationonrealartifactscanassessthegeometricshapebychangingviewing
directionandshadowcastingusingalightsource. Thesurfacefinishcanhelpwithdistinguishing
scratchesandwedges. However,duetotheverysmallwedges,mostgenuinecuneiformartifactsare
notaccessibletomanual,damage-freemeasurements. Additionally,non-uniformsurfacetexturesare
anissueduringgeometricanalysis.
Collation of photographic reproductions, on the other hand, enables non-depth related
measurements using image magnification, while the fixed camera position and lighting setup
complicatesdepthrelatedmeasurements. Rudimentarydepthperceptionissolelypossiblebasedona
staticshadow-castinglightsetup. Theeffectsofthelightingsetupare,however,inmanycasesnot
separablefromeffectscausedbysurfacetexturecolorandtone-mappingorexposurerelatedissuesof
theutilizedphotographicsetup.
Asopposedtothetwopreviously-mentionedcollationmedia,collationon3Dscannedmodels
providesdamagefreeaccesstoprecise3Dmeasurementsincludingdepthvalues,whichisespecially
importantascuneiformscriptisathree-dimensionaltypeofscript. Asidefromthat,3Dmodelscan
beviewedfromalldirectionsandwitharbitrarylightingsetups,whichalsoincludeshighpossible
levelsofmagnification. Anadditionaladvantageistheseparabilityofgeometryandsurfacetexture.
Thisallowstheexaminationofgeometricfeatureswithoutobstructionsfromnon-uniformsurface
colors.Regardingcollaborativecollationtasks,3Dscannedcuneiformdatabenefits,interalia,fromdata
mobility, reproducible viewing conditions and the conservation of the artifact state. A suitable
3Dframeworkshouldthereforeprovideinteractivevisualization,flexibleviewportnavigationand
lightmanipulation,featureenhancedrenderingofuntexturednon-2-manifoldmodelsandprecise
measurementtools,aswellasfeaturesforcollaborativeresearch.
Table 1 shows a comparison of collation accessibility features for real artifacts, photographic
reproductionsand3Dscans. Inthiscontext,representationcompletenessratesthepossibleamount
ofdataofagoodreproductioncomparedtotheoriginalartifactatthetimeofitscreation. Itisan
importantfactthatatagiventimetheoriginalartifactdoesnotnecessarilycontainthemostcomplete
Informatics 2017,4,44 4of16
data as corrosion based damage may have degraded the quality of the original artifact after the
fabricationofananalogordigitalreproduction. Thismayrenderanoldphotographyora3Dscan
themostcompletedatasourceeventhoughphotographsmaybeofbadqualityand3Dscansmay
onlycoveranincompletesubsetoftheartifactsurface. ThedatasizesinTable1refertotypicaldata
setsizesontheHethitologie-PortalMainz[3]. Itisworthmentioningthatmeasurementprecisionon
photographsisusuallysuperiortotheoriginalartifactbecauseofthepossibleimagemagnification.
However,imagesareaffectedbyprojectivedistortions,exposureandlightingissuesaswellasvisually
non-separabletexturecolors.
Table1.Collationaccessibilityfeaturesondifferenttypesoforiginalmedia.
Aspect RealObject Photography 3DScan
lighting variable fixed variable
viewingdirection variable fixed variable
measurements x,y x,y x,y,depth
manualmeasurementprecision − + ++
separabletexturecolors (cid:55) (cid:55) (cid:51)
accessibility realworldfixedlocation web-based web-based
datasize realobject medium large
representationcompleteness corrosiondependent + +
3. MethodsforWeb-BasedVisualizationof3DCuneiformData
This section will present an interactive web-based approach for the analysis of 3D scanned
cuneiform data, to address the use case of cuneiform fragment collation, described in Section 2.
TheresultwillusetheNexusformatinaThree.js-basedvisualizationenvironmentextendedwithhigh
qualityvisualizationmethodsandapplicationscenariospecifictoolsforcuneiformcollationtasksin
theHethitologie-PortalMainz. Startingwithareviewofrelatedwork,thissectionwillthereforedetail
thecomponentsofaViewerconceptasshowninFigure2. Section3.2willdiscusscuneiformdata
characteristicstomotivatetherequireddatapreparationprocess. Theemployedvisualizationmethods
willbepresentedinSection3.3,followedbyadescriptionofthedesignofasuitableuserinterfacein
Section3.4. AspectsoftheintegrationontheserverwillbecoveredinSection4.
Figure2.Outlineoftheviewerconcept.Thedatapreparationstepextendsthescandataandperforms
theconversiontothenexusformat. Theserverhoststhenexusfiles,deliversobstructeddataand
performsaccesscontrol.TheWebGLclientisresponsiblevor3Dvisualizationandprovidingauser
interface. Sharingviewsandscreenshotsaremanualcollaborativefeaturesforexchangebetween
multipleWebGLclients.
3.1. RelatedWork
An easy to use 3D web application should be based on a readily available and wide-spread
3D API without the need to install any third party plugins. WebGL is a JavaScript 3D API with
native integration in all modern browsers that satisfies these needs. Although WebGL 1.0 [4] is
based on OpenGL ES 2.0 [5], some features widely used in desktop OpenGL environments are
Informatics 2017,4,44 5of16
only available in the form of WebGL extensions with varying degree of support by the browsers.
Especiallymobiledevicebrowserslacksomewidelyusedextensionsbecauseofhardwarerelated
incompatibilities. TofacilitatemanybasicandreoccurringWebGLtasks,likeloadingshadersand
3D models, the use of high level WebGL libraries is common. In particular, open source libraries
likeThree.js[6]orSpiderGL[7],asopposedtocommercialandmoreplatformorientedprojectslike
Sketchfab[8]andUnity[9],arewellsuitedforintegratingcustomframeworkdesignsandopensource
licensedalgorithms.
AlthoughmostWebGLlibrariesofferavarietyofmethodstoloadalargeamountof3Dgeometry
file formats, many of the existing formats are not well suited for web-based loading of large 3D
models. Thismotivatestheuseofweb-optimized3Dformatsthatofferfeatureslikedatastreaming,
geometry compression and progressive model loading. Some noteworthy formats in this regard
include X3D [10] and XML3D [11] as human readable text based formats and the X3DOM Binary
Format[12]thatexternalizesthemeshdatainformofdirectlyGPUcompatiblebinaryblobswhile
stillcontainingtextbaseddescriptions. TheOpenCTMformat[13]addressesthelargedatasizeswith
entropyreductionandLZMAentropyencoding,which,asasideeffect,addsadecodingoverheadon
theclientside. TheWebGL-Loaderformat[14]combinespost-transformvertexcacheoptimization[15]
forfasterrenderingwithsophisticateddelta-encodingandquantizationtooutputaUTF-8encoded
data stream that additionally benefits from the GZIP compression of a web-server. This results in
acomparativelyhighcompressioncombinedwithverylowdecompressiontimeswhencompared
to X3D and OpenCTM [16]. X3DDOM was further extended by the POP Buffer format [17] that
employsahierarchicalquantization-basedcompressionscheme,whichenablesprogressivemodel
loading. TheNexusformat[18,19],whichwillbedetailedinSection3.2,introducesahierarchicalmesh
reductionwithfastdecodingandviewdependentrendering,enablingaprogressivevisualizationof
verylargemodels. Inthecontextof3Dscans,anadditionalusefulfeatureofthenexusformatisits
capabilitytohandlenon-2-manifoldmeshes.
Overthelastyears,various3Dframeworksfortheonlinepresentationofculturalheritageartifacts
havebeenintroduced,someofwhicharelinkedtothepublicationofscanningprojectsinmuseum
collections.SomeexamplesinthisregardwithanexclusivefocusonmusealpresentationaretheBritish
Museum,whichresortstopublishingsmallpartsoftheircollectionviaSketchfab[20],theSmithonian
InstitutionwiththeMemento-based[21](nowAutodeskRemake)SmithonianX3Dframework[22]
or the large collection of 3D scanned Vietnamese statues and architectural artifacts of VR3D [23].
Asidefromtheselargescaleprojects,therearealsofreeculturalheritagepresentationframeworkslike
3DHOP[24],whichcancombine3Dandmultimediacontentformuseumandteachingpurposesand
whichusetheNexusdataformat[18,19]. Commonfeaturesofmanyapplicationsofthesecultural
heritageviewersaretheuseoflowtomediumresolutionmodelscombinedwithmediumtohigh
resolutiontexturesandatimeconsuming,elaboratemodelpreparationregardingthecorrectionof
scanningissues.
Relatedworkdirectlyassociatedwithanalysisandpresentationof3Dcuneiformcontentincludes
theintegralinvariantbasedextractionofcuneiformsignsas2DvectordrawingsofMaraetal. [25,26]
integrated into the GigaMesh framework and the cuneiform wedge extraction presented by the
authors[27]integratedintotheCuneiformAnalyserframework. However,bothsolutionsaretargeted
atofflineusageduetolargedatasetsandcomputationallyexpensivegeometryprocessing. In2017,
theHilprechtSammlungJena[28]startedpublishingtheirlargecollectionofcuneiformtabletsonline,
takingadvantageofowningtheoriginalartifactsforallpublisheddatasetsincludingthecopyright.
Accessisprovidedbyofferingdownloadabledatasetsandbyloadingtheuncompresseddatasetsinto
MeshLabJS[29],aweb-baseddescendantofMeshLab[30],apopular,freemesheditingandconversion
solution. Alsoin2017,Collinsetal. launchedtheVirtualCuneiformTabletReconstructionProject[31],
whichincludesabasiconlineviewerformanuallyandautomaticallyjoiningcuneiformfragments.
Whileprovidingaworkingtoolforjoininglowresolutionfragmentswithfracturefaces,thismethod
wouldnotworkonincompletehighresolutiondatasetswithmissingfracturefaces. Connectedwith
Informatics 2017,4,44 6of16
thiswork,Woolleyetal.evaluatedmethodsforthedevelopmentofacollaborativevirtualenvironment
for3Dreconstructionofcuneiformtablets[32]. Thisinterestingstudyisfocusedontheanalysisof
interactionmodalitiesandusergroupsregardingtheeffectivenessofthecollaboration,whichisan
importantaspectofsystemdesign.
3.2. 3DCuneiformDataCharacteristicsandPreparation
Most3Dscannedcuneiformdatasetscanbeclassifiedintotwomaincategoriesoftentiedtothe
originofthedata. Thefirstcategoryismostlycreatedbymuseumsdigitizingtheirowninventory,
whichpredominantlyproducecomplete,highqualitydatasetsincludingtextureinformation. Thisis
possibleduetounlimitedartifactaccessandemployingsignificantamountsoftimeforscanningand
post processing the data. A second type of datasets is mainly produced by scientists with limited
accesstoartifactsownedbyforeigninstitutions. Thiscansignificantlychangetheprioritiesduring
data acquisition. For metrological analysis of cuneiform script, as targeted in the BMBF project
‘3D-JoinsundSchriftmetrologie’,capturingthescriptcoveredpartsoftheartifactswithasufficient
resolution and partial coverage of fracture faces was combined with the target to scan as many
fragmentsaspossibleduringtheavailabletimeslotswithaccesstothefragments. Inthisprocess,
thegeometricalcompletenessofnon-scriptcoveredareasandtheacquisitionoftexturedatawasonly
asecondaryobjective. Therefore,theresultingscansoftencontainunscannedsurfaceareas,holesand
non-2-manifoldgeometry. Asuitablegeometrydataformatmustbeabletodealwiththesedeficiencies
andbeabletohandleandstreamlargegeometriesaswellassatisfythephilologicalrequirements
for collation tasks as described in Section 2. In addition to this, the data format should allow the
deploymentofmethodstoobstructfullqualitymodelextractiontoacertainextentwhenneededfor
legalreasons.
Aninteractiveweb-basedscenariorequiresanoptimizationofthelatencyforviewinga3Dmodel
while the scientific usability requires a minimum model quality to enable precise measurements.
In view of the typical data sizes of cuneiform models, ranging from several megabytes to several
gigabytes,andthecurrentspeedofinternetconnections,thisimpliestheuseofstreamingandstateof
theartcompressiontechniques. Anotherimportantaspectisclientsidedecodingtime,whichmustbe
lowenoughtonotoutweighthereductionoftransfertimegainedbydatacompression.
OfthemethodsmentionedinSection3.1,theNexusformat[18,19]wastheonlyonetofeature
highcompressionrates,progressiveandlocalmodelloadingandthepossibilitytoadaptthelevelof
detailtolesscapablehardware. Inadditiontothis,thenexusformatiswellextendabletocustomdata
characteristicsandfeaturesawelldocumentedopensourcetoolsetforcreationandmodification.
TheNexusformat[18,19]isaprogressiveformatthatusesapatch-basedmultiresolutionstructure
where the patch geometry is reduced and compressed independently. Patches of neighboring
resolutionlevelsmustnotsharecommonmeshborderstoachieveauniformreductionbythequadric
simplificationalgorithmovermultipleresolutionlevels. Asuitablevolumepartitioniscreatedby
usingakd-treewithanalternatingsplitratio. Thegeometryencodingforindividualpatchesusesa
rule-basedconnectivitycompressionandaquantization-basedgeometryandattributecompression
withentropyencoding. Withoutlossycompression,generatedNexusfilesfortypicalcuneiformscans
arearound1.7timesthesizeoftheoriginalmodel,whichresultsfromthegeneratedmulti-resolution
data. Lossycompressionrates,however,dependonthequantizationsettings. Ausablecompression
ratearound4.1canbeachievedforcuneiformdata,aswillbeshowninSection5. Perceivedloadings
times are very fast, as a low resolution approximation from the highest level of the compression
hierarchyisshownalmostimmediately. Themodelisthenprogressivelyrefinedbasedonapredefined
drawbudget,amaximumcachesizeandaviewdependenttargeterror.
On3Dscannedcuneiformdata,theincrementalpatch-basedgeometryreductionofthenexus
formatisanimportantaspectforproducinghighqualitylowresolutionmeshes. Thisresultsfromthe
factthatmanymeshsimplificationmethods,liketheusedquadricedgedecimation,tendtocreate
meshesofpoorqualitywhenthetargetedgelengthismuchsmallerthantheaverageedgelengthofthe
Informatics 2017,4,44 7of16
originalmeshandtheoriginalmeshisnon-2-manifold. Non-2-manifoldnessis,however,afrequent
propertyofmany3Dscannedcuneiformdatasets. Asanimportantfactorforlesscapablehardware,
the above mentioned possibility to define a draw budget allows to limit the amount of graphics
memoryusedduringmodelrendering. Thecachesizeandadefinitionofaviewdependenttarget
error,ontheotherhand,allowtolimitthemainmemoryconsumptionandthevisuallevelofdetail.
Thisenablestheviewingofdatasetsonhardwarethat,otherwise,wouldnotbecapableofvisualizing
datasetsofthissize.
Beforethescandatacanbeconvertedintothenexusformat,thevertexdatahastobeextended
with the ambient occlusion and curvature information required for visualization as described in
the following section. The mesh data produced by the used scanner software [33] only includes
monochromaticcolorinformation. Therefore,theadditionaldatacanbecolorcodedandstoredin
the,uptothispoint,unusedcolorchannels. Asapositivesideeffect,thisavoidstheintroductionof
additionaldatachannelsintothenexusformat. Also,thedata-enrichedintermediateply-filesremain
readablebymoststandardmeshprocessingsoftwarelikeMeshLab[34].
3.3. VisualizationMethods
A scientific visualization environment for 3D cuneiform data has to address several basic
requirements,themostimportantofwhichisahighqualityvisualizationoflarge,highresolution,
untexturedmeshes. Thisresultsfromthefactthatmany3Dscannedcuneiformdatasetsrepresent
damagedcuneiformfragmentswithsubtledetails. Dependingonthescanningmethods,manyofthe
resulting3Dmodelsareuntextured,lackcolorinformationorincludecolorinformationthatexertsa
negativeinfluenceonthevisualperceptibilityofgeometricaldetails. Therefore,methodsshouldbe
usedthatoptimizethevisualspatialrepresentationofsubtledetailswhileatthesametimeproviding
arealistic,highqualityvisualizationofthe3Dmodels. Therenderingiscomplementedbytoolsfor
takingmeasurements,usercontrollablelighting,astylizedautographymodeandanintuitiveuser
interfaceincludingviewportnavigationandparametercontrols.
As the set of visualization methods used in previous work of the authors [27] has proven its
usefulnessforofflineanalysisofcuneiformdatasets,theframeworkusesacombinationofambient
occlusion, radiance scaling and lit sphere shading to render the datasets. Vertex-based ambient
occlusion[35]approximatestheselfshadowingpropertiesofthesurfacethatcansignificantlyimprove
thedepthperceptionofsurfacestructureswhenusedtocontroltheluminosityofthesurfaceshading.
Although a real time approximation for ambient occlusion can be computed in screen space [36],
this framework uses a precomputed, view-independent, gpu-accelerated ambient occlusion term,
asdescribedin[37],forimprovedaccuracy.
Realistic shading is achieved by using precomputed lighting in the form of lit-spheres. In a
preparationstep,anorthographicimageofashadedsphere,includingacomplexsurfacematerialan
arbitrarynumberoflightsandareflectiveenvironment,isrenderedintoasquaretexture. Theshading
isthencomputedbyprojectingthenormalizedsurfacenormalorthereflectedviewdirectiononto
thelit-spheretexture. Theprocessofpre-renderingatexturecanbereplacedbytakingphotographs
ofsphericalobjectsinrealworldmaterialandlightingconditionsandcroppingtheresultingimage
totheextendsofthesphere. Itisworthmentioningthattheresultinglit-spheretexturesmustnot
containlocalsurfacefeaturesasopposedtocomponentsoftheshadingfunctionlikeglobalmaterial
color,lightingandreflections. Thereforelocalsurfacefeaturesmustberemovedbyappropriateimage
processingmethodlikesmoothingorcloning.
To address cuneiform script analysis specifically, several types of lit-sphere textures were
generated to resemble clay and metal type materials. This was realized by taking photos of clay
spheres,includingnecessarypost-processing,aswellasbyrenderingsphereswithcomplexshaders
usingahighqualityofflinerenderer.Theassociatedlightingsetupswerechosentoresemblefrequently
usedlightingsituationsthatcanbefoundintraditionalphotographiccuneiformreproductionsto
enhancewedgevisualization. Anespeciallycommonlightingsetupincludesadirectionallightsource
Informatics 2017,4,44 8of16
positionedwitha45degreeangleatthetopleftofafragmentasthislightpositioncastsadvantageous
shadowsonmanycuneiformwedgetypes.
Aselectionofthreetypesoflit-spheresusedintheframeworkandtheresultingvisualappearance
includingthelit-spheretexturesisshowninFigure3a–c. Thedepictedlit-spheretextureshavebeen
additionallyprocessedtocounteractrenderingartifactsbyextrapolatingthecoloratthesphereborder
tothesquaretextureborder.
(a) (b) (c)
Figure3.Asetofdifferentmaterials,withthecorrespondingsquarelit-spheretextureshowninthe
bottomright,includingdiffuseclay(a),glossyclay(b)andpolishedgold(c).
To visually enhance geometrical details on the scanned surfaces, the radiance scaling
technique[38]isemployedthatcorrelatessurfacevariationstoaspectsoftheshadingfunction. Asthe
namesuggests,radiancescalingadjuststhereflectedlightintensitydependingonthesurfacecurvature
andmaterialproperties. Thisisrealizedbyapplyinganindependentmonotonicscalingfunctionto
theoriginalBidirectionalReflectanceDistributionFunction. Asdetailedin[38],itsdesigndependson
aninvariancepointα,inordernottoaffectplanarregions,andascalingmagnitudeγ. Thescaling
functionisparameterizedusinganormalizedreflectanceandanormalizedcurvature. InanOpenGL-
andGLSL-basedenvironmentthereflectancecanbeobtainedbycomputingthelengthofthecolor
vectorresultingfromanarbitraryshadingfunctionlikephongshading,lit-sphereshadingorashading
functioncomponentsuchasspecularityorambientocclusion. Thenormalizedcurvatureisobtained
asahyperbolictangentmappedmeancurvature,asindescribedin[39].
Inthiswork,thescalingfunctionisparameterizedwithascalinginvariantpoint α = 0.1and
ausercontrollablescalingmagnitudeγ ∈ [0,2]. Thescalingisappliedonlytothecolortermofthe
shading function. Additional components, like the ambient occlusion, are applied independently
astheyhavebeenfoundtoexertanegativeinfluenceontherenderingofgeometricdetailswithin
cuneiformwedges. Asboththenormal-basedgradientandcurvaturecalculationsrequireaccessinga
localneighborhoodofapixeltobeshaded,adeferredshadingarchitectureisused.
Figure4a–dshowstheamountofvisualimprovementsasopposedtosimplephong-shading
onanexemplarycuneiformfragment. Theeffectsofdetailenhancementusingradiancescalingon
cuneiformwedgesisshowninFigure5. Themethodhasproventobeparticularlyusefulonabraded
andcorrodedsurfaceareaswithsubtledetails.
Informatics 2017,4,44 9of16
(a) (b) (c) (d)
Figure4. Renderingimprovementoverphong-shading(a)withlit-sphereshading(b), additional
ambient occlusion (c) and additional ambient occlusion and radiance scaling (d) on cuneiform
fragmentE971.
(a) (b) (c)
Figure5.SectionofcuneiformfragmentE971without(a)andwithradiancescaling(b)andastylized
rendering(c)ofthesamesectioninautographymode.
3.4. UserInterfaceDesignforCuneiformAnalysis
Asidefromtheaspectsdirectlytiedtotransferandrenderingofthe3Dmodels,theuserinterface
of the viewer is designed with two main objectives in mind. The interface should provide an
intuitiveaccesstothemodels,whileatthesametimeaddressingmostoftherequirementsofcollation
relatedtasks.
Philologistsoftenanalyzegeometricdetailsonrealartifactsbyvaryingthepositionofadirected
lightsource. Therefore,theframeworkfeaturesalightdirectionmodethatcanbeusedtodirectly
controlthedirectionoflightingwiththemousepointeronavirtualsphere. Whilephong-shading
offersunrestrictedcontrolofthelightposition,thelightingpositionoflit-sphereshadingcanonlybe
influencedbyrotatingthelitspheretexturesintheimageplane. Alsoinspiredbytraditionalcuneiform
analysismethods,theframeworkintegratesanautographymode,whichrendersastylizedblackand
white image of the geometry, as shown in Figure 5c, based on precomputed maximum curvature
values. Thisrenderingstyleresembleshanddrawnautographiesofcuneiformfragments,liketheone
depictedinFigure1a,whichcanbehelpfulduringthetranscriptionprocess.
Astheorientationofscannedcuneiformfragmentsandthecontainedtextsin3Dspaceisnot
known a priori, the viewport navigation is realized using the quaternion-based arcball method.
This allows turning the models in any desired direction without experiencing gimbal lock related
sideeffectsofeulerrotationand,asopposedtothetrackballmethod,anintuitivewayofrotatingthe
viewportintheimageplane.Thelatterisespeciallyimportanttoaligncuneiformtextlineshorizontally
intheviewport. Theinitialzoomiscoupledwiththesizeofthegeometrytoalwaysdisplaythefull
Informatics 2017,4,44 10of16
3Dmodelwhenopeninganewdataset. Toavoiddepthbufferissueswiththerequiredwidezoom
range, the framework uses an optimized logarithmic depth buffer implementation [40] like in the
OuterraEngine.
Toassistwithcomplexmeasurementtasks,theframeworkintegratesameasurementmodeand
anorthographiccameramodewithadynamicscaleobject. Themeasurementmodeallowstoposition
multiplemeasurementtapesonthesurfaceoftheobject,asshowninFigure6. Themeasureddistances
aredisplayedastextoverlaysonthescreenandinformofamillimetersizedstripepatternonthe
measurement tape object. It has to be noted that the measurement mode works only on scanned
objects with a known system of measurement. Although most scans are measured in millimeter
scale,scanscreatedusingphotometricreconstructiontechniques,likelightdomes[41],mayexhibit
significantdistortionsthatpreventprecisemeasurements. Muchlikeinmanytraditionalphotographic
reproductions of cuneiform fragments, the scale object, as can be seen in Figure 6, consists of an
alternatingdualrowpatternofblackandwhitesquares. Thescaleofthesquaresandanassociated
measurementunitisadjusteddynamicallyastheuserzoomstheviewport. Thisway,measurements
canbetakeneveninprintedscreenshots. Tofurthersupportcollaborativeworkonthe3Dmodels,
the viewer supports the export of links to reopen models with the current camera configuration,
which is also useful for inventory related tasks. For instance, occurrences of sign forms could be
directlyreferencedfroma3Dsigninventorydatabase.
Figure6. Thegraphicaluserinterfaceoftheframework,includingquickaccessbuttonsonthetop
left,thedat.GUI[42]basedconfigurationoftherenderingpipelineonthetopright,thedynamicscale
objectatthebottomleftandseveralmeasurementtapesonthemodel.
4. IntegrationintotheHethitologie-PortalMainz
Theweb-based3DapproachisintegratedintotheHethitologie-PortalMainz,theleadingonline
researchresourceinthefieldofHittitology. Onecentralfeatureoftheportalistheconcordanceof
Hittitecuneiformtablets. Thisonline-accessibledatabaseincludesalargesearchablecatalogofknown
cuneiformtabletslinkedwithaphotographiccollection,join-sketches,transcriptions,transliterations,
publication and provenience data. To ensure an intuitive data integration, the 3D models can
be accessed directly from the concordance query results much like the traditional photographic
reproductions. The possibility of accessing a 3D model depends not only on the existence of a
respective 3D scan but also on the publication state of the corresponding cuneiform manuscript,
whichonlypermitsaccesstoalreadypublishedmanuscripts. Figure7showsrepresentativequery
resultintheconcordance,listingcuneiformdocumentswhichmatchthesearchcriteria. Beginningon
thelefthandside,thecolumn“Inventarnummer”containstheinventorynumber,aniconindicatingthe
existenceofajoinlayoutsketch,arediconindicatingtheexistenceofphotographicreproductionsand
thetext“3D”ifaninteractive3Dscanisavailable. Layoutsketches,2Dreproductionsand3Dscansare
directlylinkedfromtheseelements. Theremainingcolumns,ifavailable,includeapublicationnumber
(withlinkedautographyimages),anindextotheCataloguedesTextesHittites(agenreclassificationof
Hittitedocuments),aplaceofdiscovery,adatingandasetofannotationsregardingmanuscriptcontent