20.500.12556/RUL-183762 Vpliv generativne umetne inteligence na fotografski delokrog Influence of generative artificial intelligence on photographic workflow V diplomskem delu je predstavljeno področje generativne umetne inteligence v procesu postprodukcije fotografij. Na podlagi različnih motivov fotografij je predstavljen celoten proces postprodukcije ter izvedena primerjava med klasično ročno obdelavo in sodobnimi pristopi, ki temeljijo na uporabi umetne inteligence. V teoretičnem delu je najprej predstavljen razvoj fotografije in postprodukcije ter pregled obstoječih tehnologij, orodij in programov. Opisana so orodja za ročno obdelavo fotografij, kot sta Adobe Photoshop in Adobe Lightroom, ter sodobna orodja, ki vključujejo generativno umetno inteligenco, kot so Adobe Firefly, ChatGPT in Gemini. V eksperimentalnem delu je bila izvedena primerjava različnih načinov obdelave fotografij glede na časovno učinkovitost, število korakov in končni vizualni rezultat. Obdelane fotografije so bile nato analizirane tudi s pomočjo ankete, v kateri so udeleženci ocenjevali njihovo vizualno privlačnost. Na podlagi analize rezultatov smo ugotovili, kateri pristopi omogočajo hitrejšo obdelavo ter kateri dosegajo najboljše rezultate z vidika zaznave uporabnikov. Rezultati so pokazali, da pristopi, ki temeljijo na generativni umetni inteligenci, omogočajo hitrejšo in bolj učinkovito obdelavo fotografij ter poenostavitev delovnega procesa. Kljub temu je bilo ugotovljeno, da ročna obdelava v določenih primerih dosega višjo raven kakovosti, predvsem z vidika natančnosti in nadzora nad končno podobo. Na podlagi ugotovitev je bilo zaključeno, da generativna umetna inteligenca predstavlja pomembno podporno orodje v postprodukciji fotografij, vendar ne more v celoti nadomestiti vloge fotografa. The thesis presented the field of generative artificial intelligence in the process of photographic post-production. Based on various photographic motifs, the entire post-production process was outlined and a comparison between traditional manual editing and contemporary approaches based on artificial intelligence was conducted. In the theoretical part, the development of photography and post-production was presented, together with an overview of existing technologies, tools, and software. Manual editing tools such as Adobe Photoshop and Adobe Lightroom were described, as well as modern tools incorporating generative artificial intelligence, including Adobe Firefly, ChatGPT, and Gemini. In the experimental part, a comparison of different methods of photographic post-processing was carried out with regard to time efficiency, the number of processing steps, and the final visual result. The processed photographs were further analysed through a survey in which participants evaluated their visual appeal. Based on the analysis of the results, it was determined which approaches enable faster processing and which achieve better results in terms of user perception. The results showed that approaches based on generative artificial intelligence enable faster and more efficient photo editing and simplify the workflow. However, it was found that manual editing in certain cases achieves a higher level of quality, particularly in terms of precision and control over the final image. Based on the findings, it was concluded that generative artificial intelligence represents an important supportive tool in photographic post-production, but cannot fully replace the role of the photographer. fotografija fotografski delokrog generativna umetna inteligenca obdelava fotografij postprodukcija vizualna zaznava. generative artificial intelligence image processing photographic workflow photography post-production visual perception true false false Slovenski jezik Angleški jezik Diplomsko delo/naloga 2026-06-18 09:01:14 2026-06-18 09:01:18 2026-06-19 04:14:12 0000-00-00 00:00:00 2026 0 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 1970-01-01 114054 29355.pdf 29355.pdf 1 56D16DA3426F026BE9332E9A1266BC5B dc08d7c33e86717491b8bcc9fbe9321d965ba120c4f7acb0fc4870f7297377ff 7319f870-6ae3-11f1-9b0d-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=235919 Naravoslovnotehniška fakulteta 0 0 0