GENERATIVE ADVERSARIAL NETWORKS (GAN) DALAM FOTOGRAFI: MENCIPTAKAN IMAJI DARI NOL
Abstract
Generative Adversarial Networks (GAN) merupakan salah satu terobosan penting dalam kecerdasan buatan yang telah memberikan dampak signifikan pada dunia fotografi. Teknologi ini memungkinkan pembuatan imaji foto realistis dari data acak, selanjutnya menciptakan peluang baru dalam produksi foto. Penelitian ini mengeksplorasi sejumlah hasil penelitian tentang penerapan GAN dalam fotografi serta mengkaji implikasi estetika dan etika yang muncul seiring penggunaannya. Metode yang digunakan adalah pendekatan kualitatif dengan studi literatur, mengumpulkan data dari berbagai artikel ilmiah, buku, dan publikasi akademis yang berfokus pada GAN dan aplikasinya dalam pembuatan imaji foto. Hasil penelitian menunjukkan bahwa GAN memungkinkan kreasi imaji foto baru yang sebelumnya tidak mungkin dilakukan, serta memberikan kemampuan untuk memodifikasi secara kreatif. Namun, penerapan teknologi ini juga memunculkan tantangan terkait keaslian dan kredibilitas foto yang dihasilkannya, terutama dalam konteks deepfake dan manipulasi. Selain itu, terdapat kekhawatiran mengenai dampak penggunaan GAN terhadap persepsi publik mengenai otentisitasnya. Penelitian ini menyimpulkan bahwa GAN memberikan kontribusi besar dalam pengembangan estetika fotografi, namun diperlukan regulasi dan perhatian lebih terhadap aspek etika untuk menjaga integritas seni fotografi di era digital.
Generative Adversarial Networks (GAN) in Photography: Creating Images from Scratch. Generative Adversarial Networks (GAN) represent a significant breakthrough in artificial intelligence that has profoundly impacted the world of photography. This technology enables the creation of photorealistic images from random data, thereby opening up new opportunities in photo production. This study explores a range of research on applying GANs in photography and examines the aesthetic and ethical implications that arise from their use. The method employed is a qualitative approach with a literature review, gathering data from various scientific articles, books, and academic publications focused on GANs and their applications in image creation. The findings indicate that GANs enable the creation of new photographic images that were previously impossible while also providing the ability to creatively modify pictures. However, the application of this technology also raises challenges regarding the authenticity and credibility of the photos it generates, particularly in the context of deepfakes and manipulation. Furthermore, there are concerns about the impact of GAN usage on public perception of its authenticity. This research concludes that GANs significantly contribute to the development of photographic aesthetics. Still, regulation and greater attention to ethical aspects are needed to maintain the integrity of photographic art in the digital age.
Keywords
Full Text:
PDF (Bahasa Indonesia)References
Brock, Andrew, et al. “Large Scale GAN Training for High Fidelity Natural Image Synthesis.” International Conference on Learning Representations (ICLR), 2019, https://openreview.net/forum?id=B1xsqj09Fm.
Chen, Jifeng, and Vladlen Koltun. “Photographic Image Synthesis with Cascaded Refinement Networks.” Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2017, pp. 1511–20, https://doi.org/10.1109/ICCV.2017.171.
Chesney, Robert, and Danielle Citron. “Deepfakes and the New Disinformation War: The Coming Age of Post-Truth Geopolitics.” Foreign Affairs, 2019, https://www.foreignaffairs.com/articles/world/2018-12-11/deepfakes-and-new-disinformation-war.
Elgammal, Ahmed, et al. “CAN: Creative Adversarial Networks, Generating ‘Art’ by Learning about Styles and Deviating from Style Norms.” ArXiv Preprint ArXiv:1706.07068, 2017, https://arxiv.org/abs/1706.07068.
Gervais, Daniel J. “The Machine as Author.” Iowa Law Review, vol. 105, no. 5, 2019, pp. 2053–106, https://ilr.law.uiowa.edu/print/volume-105-issue-5/the-machine-as-author/.
Goodfellow, Ian, et al. “Generative Adversarial Nets.” Advances in Neural Information Processing Systems (NeurIPS), vol. 27, 2014, pp. 2672–80, https://papers.nips.cc/paper/2014/file/5ca3e9b122f61f8f06494c97b1afccf3-Paper.pdf.
Karras, Tero, Samuli Laine, et al. “A Style-Based Generator Architecture for Generative Adversarial Networks.” Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019, pp. 4401–10, https://doi.org/10.1109/CVPR.2019.00453.
Karras, Tero, Miika Aittala, et al. “Training Generative Adversarial Networks with Limited Data.” ArXiv Preprint ArXiv:2006.06676, 2020, https://arxiv.org/abs/2006.06676.
Korshunov, Pavel, and S’ebastien Marcel. “Deepfakes: A New Threat to Face Recognition? Assessment and Detection.” ArXiv Preprint ArXiv:1812.08685, 2018, https://arxiv.org/abs/1812.08685.
Ledig, Christian, et al. “Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network.” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 4681–90, https://doi.org/10.1109/CVPR.2017.703.
Radford, Alec, et al. “Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks.” ArXiv Preprint ArXiv:1511.06434, 2016, https://arxiv.org/abs/1511.06434.
Ramesh, Aditya, et al. “Zero-Shot Text-to-Image Generation.” ArXiv Preprint ArXiv:2102.12092, 2021, https://arxiv.org/abs/2102.12092.
Ronald, Regina Octavia, and Tri Karyono. “Perubahan Lanskap Kreatif dan Dialog Etika dalam Fotografi: Pengaruh Kecerdasan Buatan.” SPECTA, vol. 8, no. 1, May 2024, pp. 15–34.
Schwartz, R., et al. “Green AI.” Communications of the ACM, vol. 63, no. 12, 2020, pp. 54–63.
Yeh, Raymond A., et al. “Semantic Image Inpainting with Deep Generative Models.” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 5485–93, https://doi.org/10.1109/CVPR.2017.583.
Yudisetyanto, Raynald Alfian, and Achmad Taufik Firmansyah. “Fotografi pada Era Disrupsi: Artificial Intelligence sebagai Referensi dalam Mengembangkan Ide Kreatif Fotografi.” SPECTA, vol. 8, no. 1, May 2024, pp. 35–44.
Zhu, Jun-Yan, et al. “Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks.” Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2017, pp. 2242–51, https://doi.org/10.1109/ICCV.2017.244.
DOI: https://doi.org/10.24821/specta.v8i2.13910
Article Metrics
Abstract view : 0 timesPDF (Bahasa Indonesia) - 0 times
Refbacks
- There are currently no refbacks.
This work is licensed under a Creative Commons Attribution 4.0 International License.