AI ASSISTED FACIAL RECONSTRUCTION OF PANJI’S LOVER CHARACTER IN PENATARAN TEMPLE’S BAS RELIEFS USING SKETCH TO IMAGE METHOD
DOI:
https://doi.org/10.58982/jadam.v4i1.605Keywords:
artificial intelligence, Panji Tales, sketch to image, generation, PenataranAbstract
Purpose: Purpose of this research is to realistically reconstruct face of Panji’s Lover character at 51st panel of Teras Pendopo structure’s bas relief in Penataran Temple complex as guide for concept art creation of Panji’s Lover character.
Research methods: This experiment consists of two steps. First is manually drawing a sketch of the character and describing her facial features. The second step is image generation by artificial intellegence employing sketch-to-image method using created sketch as base image and the description as prompt. The used artificial intelligence (AI) program is Stable Diffusion 1.5 with ControlNet add on. The used AI models are Realistic Vision 5.1 and Dreamshaper 8.
Findings: Images generated with Realistic Vision 5.1 model are closest to sketch and description of original bas relief.
Implications: The results of this experiment can’t be used as final artwork but can be guide for concept art creation for both Candra Kirana and Galuh Ajeng character in any future visual art projects based on Panji Tales stories.
Downloads
References
L. Kieven, Menelusuri Figur Bertopi Dalam Relief Candi Zaman Majapahit. Jakarta: Kepustakaan Populer Gramedia, 2014.
I. B. P. Manuaba, A. Setijowati, and P. Karyanto, “Keberadaan dan Bentuk Transformasi Cerita Panji,” LITERA: Jurnal Penelitian Bahasa, Sastra, dan Pengajarannya, vol. 12, no. Nomor 1, APRIL 2013, pp. 53–67, 2013, [Online]. Available: http://journal.uny.ac.id/index.php/litera/issue/view/221
L. Kieven, Following the Cap-Figure in Majapahit Temple Reliefs. Leiden: Brill, 2013. doi: 10.1163/9789004258655.
Joel, “Gambar Umbul: Ande Ande Lumut,” Laba-Laba Lapak. Accessed: Nov. 29, 2016. [Online]. Available: http://labalabalapak.blogspot.co.id/2013/07/ande-ande-lumut.html
R. Wibowo, Alkisah. Jakarta: Djarum Foundation, 2016.
A. Borji, Q. Ai, and S. Francisco, “Generated Faces in the Wild: Quantitative Comparison of Stable Diffusion, Midjourney and DALL-E 2,” Oct. 2022, Accessed: Oct. 28, 2023. [Online]. Available: https://arxiv.org/abs/2210.00586v2
A. Stöckl, “Evaluating a Synthetic Image Dataset Generated with Stable Diffusion,” pp. 805–818, 2023, doi: 10.1007/978-981-99-3243-6_64/COVER.
L. Lykon, “DreamShaper,” Civit.ai. Accessed: Oct. 28, 2023. [Online]. Available: https://civitai.com/models/4384/dreamshaper
SG_161222, “Realistic Vision V5.1,” Civit.ai. Accessed: Oct. 28, 2023. [Online]. Available: https://civitai.com/models/4201/realistic-vision-v51
L. Zhang, “sd-controlnet-canny,” huggingface.co. Accessed: Oct. 28, 2023. [Online]. Available: https://huggingface.co/lllyasviel/sd-controlnet-canny/tree/main
T. Andito, “PERANCANGAN KOMIK ANDHE ANDHE LUMUT BERDASARKAN RELIEF KISAH PANJI DI KOMPLEKS CANDI PENATARAN,” DeKaVe, vol. 10, no. 2, p. 49, Mar. 2017, doi: 10.24821/dkv.v10i2.1993.
“Guide to Stable Diffusion CFG scale (guidance scale) parameter | getimg.ai.” Accessed: Apr. 01, 2024. [Online]. Available: https://getimg.ai/guides/interactive-guide-to-stable-diffusion-guidance-scale-parameter
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2024 Tegar Andito
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
