| fullrecord |
[{"key": "dc.contributor.advisor", "value": "Meht\u00e4l\u00e4, Saana", "language": "", "element": "contributor", "qualifier": "advisor", "schema": "dc"}, {"key": "dc.contributor.advisor", "value": "Sepp\u00e4nen, Ville", "language": "", "element": "contributor", "qualifier": "advisor", "schema": "dc"}, {"key": "dc.contributor.author", "value": "Karppinen, Kalle", "language": "", "element": "contributor", "qualifier": "author", "schema": "dc"}, {"key": "dc.date.accessioned", "value": "2023-06-30T10:58:05Z", "language": null, "element": "date", "qualifier": "accessioned", "schema": "dc"}, {"key": "dc.date.available", "value": "2023-06-30T10:58:05Z", "language": null, "element": "date", "qualifier": "available", "schema": "dc"}, {"key": "dc.date.issued", "value": "2023", "language": "", "element": "date", "qualifier": "issued", "schema": "dc"}, {"key": "dc.identifier.uri", "value": "https://jyx.jyu.fi/handle/123456789/88162", "language": null, "element": "identifier", "qualifier": "uri", "schema": "dc"}, {"key": "dc.description.abstract", "value": "T\u00e4m\u00e4 kandidaatinty\u00f6 tutki latenteilla diffuusiomalleilla (LDM) muodostettujen kuvien soveltuvuutta fotorealistiseen tuotekuvastamiseen ja ihmisten n\u00e4kemyksi\u00e4 LDM:ll\u00e4 generoiduista tuotekuvista. Lis\u00e4ksi tutkimuksessa pohdittiin synteettiseen kuvangenerointiin liittyvi\u00e4 eettisi\u00e4 n\u00e4k\u00f6kulmia UNESCO:n ja ECCOLA-viitekehyksien perusteella. LDM on koneoppimismalli, mik\u00e4 sis\u00e4lt\u00e4\u00e4 pakattua tietoa kuvan keskeisist\u00e4 piirteist\u00e4, jota voidaan my\u00f6hemmin k\u00e4ytt\u00e4\u00e4 uusien kuvien muodostamisessa. Fotorealismilla tarkoitetaan kuvan pyrkimyst\u00e4 olla oikean valokuvan n\u00e4k\u00f6inen. Ty\u00f6ss\u00e4 ker\u00e4ttiin tietoa tuotekuvastamisesta erityisesti markkinointiviestinn\u00e4llisi\u00e4 kanavia varten, ja luotiin kirjallisuuskatsauksen muodossa kuvauksia kirjoitushetkell\u00e4 laadukkaimpia tuloksia tuottavista tekniikoista ja toimintamalleista. Tuotetun sis\u00e4ll\u00f6n laadullisen onnistumisen ja tekniikoiden k\u00e4ytt\u00e4misen eettisten seikkojen tarkastelu oli my\u00f6s olennainen osa tutkimusta. Kirjallisuuskatsauksen lis\u00e4ksi tutkimuksessa ker\u00e4ttiin dataa mielipidekyselyn muodossa, jolla selvitettiin ihmisten toiveita synteettisen sis\u00e4ll\u00f6n k\u00e4ytt\u00e4misest\u00e4 markkinointiviestinn\u00e4llisen\u00e4 v\u00e4lineen\u00e4. Kyselytutkimuksella pyrittiin sitomaan tutkimusaiheen teoriaa todellisuuteen, ja todistamaan kuvangeneroimisen toimivuutta ja k\u00e4ytt\u00f6kelpoisuutta. Kysely\u00e4 ja t\u00e4t\u00e4 tutkielmaa varten luodut kuvat valmistettiin kuvaamalla reaalimaailman kohteesta nelj\u00e4ntoista kuvan kuvatietokanta, joka siirto-opetettiin olemassa olevan latentin diffuusiomallin Stable Diffusion v.2.1 non-EMA:n alkuper\u00e4isten arvojen yhteyteen hy\u00f6dynt\u00e4en suurille kielimalleille kehitetty\u00e4 LoRA (Low Ranking Adaptation of Large Language Models) menetelm\u00e4\u00e4. Kuvattavaksi kohteeksi valikoitui symmetrinen pinnatuolimalli, joka kuvattiin nelj\u00e4ss\u00e4 eri v\u00e4riss\u00e4. Kandidaatinty\u00f6n kyselytutkimus toteutettiin Google Forms mielipidekyselyn\u00e4, jossa vastaajat saivat kertoa mielipiteit\u00e4\u00e4n synteettisen sis\u00e4ll\u00f6n k\u00e4ytt\u00e4misest\u00e4 markkinointiviestinn\u00e4n v\u00e4lineen\u00e4, ja arvioida kuvien fotorealistista uskottavuutta Likert-asteikolla 1\u20135 ja vapaavalintaisten lis\u00e4kysymysten muodossa. Kyselytutkimustulokset osoittivat tekniikoiden omaavan suuren potentiaalin fotorealistisessa sis\u00e4ll\u00f6ntuottamisessa, ja niiden k\u00e4ytt\u00e4misen olevan jossain tilanteissa markkinointiviestinn\u00e4llisesti kannattavaa. Tutkimustulosten perusteella kuvasynteesill\u00e4 voidaan saavuttaa jo ainakin suhteellinen fotorealismin taso. Toisaalta ne my\u00f6s kertovat ihmisten mielipiteist\u00e4 siit\u00e4, miten teko\u00e4lytekniikoiden k\u00e4ytt\u00e4minen tulisi tehd\u00e4 avoimesti ja aihepiirin eettisi\u00e4 aspekteja miettien, jottei kuluttajia johdettaisi harhaan.", "language": "fi", "element": "description", "qualifier": "abstract", "schema": "dc"}, {"key": "dc.description.abstract", "value": "This bachelor's thesis investigated the suitability of images generated by Latent Diffusion Models (LDM) for photorealistic product representation and people's perceptions of LDM-generated product images. Additionally, the study explored ethical perspectives related to synthetic image generation based on the UNESCO and ECCOLA frameworks. LDM is a machine learning model that encapsulates compressed information about the key features of an image, which can later be used to generate new images. Photorealism refers to the image's attempt to resemble a real photograph. The thesis collected information on product representation, especially for marketing communication channels, and created literature reviews describing the techniques and methodologies that produced the highest quality results at the time of writing. Examining the qualitative success of the generated content and the ethical considerations of using the techniques was also an essential part of the research. In addition to the literature review, data was gathered through an opinion survey to explore people's desires regarding the use of synthetic content as a marketing communication tool. The survey aimed to bridge the gap between theoretical aspects of the research topic and reality, as well as to validate the effectiveness and usability of image generation. Images for the survey and this thesis were created by capturing a real-world object, a symmetric upholstered chair model, in fourteen different image variations. The image dataset was then transfer-trained into the existing Stable Diffusion v.2.1 non-EMA latent diffusion model using the LoRA (Low Ranking Adaptation of Large Language Models) method developed for large-scale language models. The bachelor's thesis survey was conducted using Google Forms, where respondents could express their opinions on using synthetic content as a marketing communication tool and evaluate the photorealistic credibility of the images on a Likert scale of 1-5, as well as through additional open-ended questions. The survey results indicated that the techniques have significant potential in photorealistic content generation, and their use can be beneficial in certain marketing communication contexts. Based on the research findings, relative photorealism levels can already be achieved through image synthesis. However, the results also highlight people's opinions on how the use of AI technologies should be conducted transparently and with consideration for ethical aspects, to avoid misleading consumers.", "language": "en", "element": "description", "qualifier": "abstract", "schema": "dc"}, {"key": "dc.description.provenance", "value": "Submitted by Miia Hakanen (mihakane@jyu.fi) on 2023-06-30T10:58:05Z\nNo. of bitstreams: 0", "language": "en", "element": "description", "qualifier": "provenance", "schema": "dc"}, {"key": "dc.description.provenance", "value": "Made available in DSpace on 2023-06-30T10:58:05Z (GMT). No. of bitstreams: 0\n Previous issue date: 2023", "language": "en", "element": "description", "qualifier": "provenance", "schema": "dc"}, {"key": "dc.format.extent", "value": "89", "language": "", "element": "format", "qualifier": "extent", "schema": "dc"}, {"key": "dc.language.iso", "value": "fin", "language": null, "element": "language", "qualifier": "iso", "schema": "dc"}, {"key": "dc.rights", "value": "In Copyright", "language": "en", "element": "rights", "qualifier": null, "schema": "dc"}, {"key": "dc.subject.other", "value": "kuvangeneroiminen", "language": "", "element": "subject", "qualifier": "other", "schema": "dc"}, {"key": "dc.subject.other", "value": "latentit diffuusiomallit", "language": "", "element": "subject", "qualifier": "other", "schema": "dc"}, {"key": "dc.title", "value": "Fotorealistinen tuotekuvastaminen latenteilla diffuusiomalleilla", "language": "", "element": "title", "qualifier": null, "schema": "dc"}, {"key": "dc.type", "value": "bachelor thesis", "language": null, "element": "type", "qualifier": null, "schema": "dc"}, {"key": "dc.identifier.urn", "value": "URN:NBN:fi:jyu-202306304307", "language": "", "element": "identifier", "qualifier": "urn", "schema": "dc"}, {"key": "dc.type.ontasot", "value": "Bachelor's thesis", "language": "en", "element": "type", "qualifier": "ontasot", "schema": "dc"}, {"key": "dc.type.ontasot", "value": "Kandidaatinty\u00f6", "language": "fi", "element": "type", "qualifier": "ontasot", "schema": "dc"}, {"key": "dc.contributor.faculty", "value": "Informaatioteknologian tiedekunta", "language": "fi", "element": "contributor", "qualifier": "faculty", "schema": "dc"}, {"key": "dc.contributor.faculty", "value": "Faculty of Information Technology", "language": "en", "element": "contributor", "qualifier": "faculty", "schema": "dc"}, {"key": "dc.contributor.department", "value": "Informaatioteknologia", "language": "fi", "element": "contributor", "qualifier": "department", "schema": "dc"}, {"key": "dc.contributor.department", "value": "Information Technology", "language": "en", "element": "contributor", "qualifier": "department", "schema": "dc"}, {"key": "dc.contributor.organization", "value": "Jyv\u00e4skyl\u00e4n yliopisto", "language": "fi", "element": "contributor", "qualifier": "organization", "schema": "dc"}, {"key": "dc.contributor.organization", "value": "University of Jyv\u00e4skyl\u00e4", "language": "en", "element": "contributor", "qualifier": "organization", "schema": "dc"}, {"key": "dc.subject.discipline", "value": "Tietoj\u00e4rjestelm\u00e4tiede", "language": "fi", "element": "subject", "qualifier": "discipline", "schema": "dc"}, {"key": "dc.subject.discipline", "value": "Information Systems Science", "language": "en", "element": "subject", "qualifier": "discipline", "schema": "dc"}, {"key": "yvv.contractresearch.funding", "value": "0", "language": "", "element": "contractresearch", "qualifier": "funding", "schema": "yvv"}, {"key": "dc.type.coar", "value": "http://purl.org/coar/resource_type/c_7a1f", "language": null, "element": "type", "qualifier": "coar", "schema": "dc"}, {"key": "dc.rights.accesslevel", "value": "openAccess", "language": null, "element": "rights", "qualifier": "accesslevel", "schema": "dc"}, {"key": "dc.type.publication", "value": "bachelorThesis", "language": null, "element": "type", "qualifier": "publication", "schema": "dc"}, {"key": "dc.subject.oppiainekoodi", "value": "601", "language": "", "element": "subject", "qualifier": "oppiainekoodi", "schema": "dc"}, {"key": "dc.subject.yso", "value": "fotorealismi", "language": null, "element": "subject", "qualifier": "yso", "schema": "dc"}, {"key": "dc.subject.yso", "value": "markkinointi", "language": null, "element": "subject", "qualifier": "yso", "schema": "dc"}, {"key": "dc.subject.yso", "value": "etiikka", "language": null, "element": "subject", "qualifier": "yso", "schema": "dc"}, {"key": "dc.subject.yso", "value": "teko\u00e4ly", "language": null, "element": "subject", "qualifier": "yso", "schema": "dc"}, {"key": "dc.subject.yso", "value": "kuvat", "language": null, "element": "subject", "qualifier": "yso", "schema": "dc"}, {"key": "dc.subject.yso", "value": "valokuvat", "language": null, "element": "subject", "qualifier": "yso", "schema": "dc"}, {"key": "dc.subject.yso", "value": "kuvamanipulaatio", "language": null, "element": "subject", "qualifier": "yso", "schema": "dc"}, {"key": "dc.rights.url", "value": "https://rightsstatements.org/page/InC/1.0/", "language": null, "element": "rights", "qualifier": "url", "schema": "dc"}]
|