Analisis Perbandingan Kinerja Metode GFPGAN dan RestoreFormer dalam Restorasi Foto Lama

Hariman Bahtiar; Amri Muliawan Nur; Almi Yulistia Alwanda; Zulkipli Zulkipli

doi:10.30591/jpit.v11i2.10291

Analisis Perbandingan Kinerja Metode GFPGAN dan RestoreFormer dalam Restorasi Foto Lama

Hariman Bahtiar, Amri Muliawan Nur, Almi Yulistia Alwanda, Zulkipli Zulkipli

Abstract

A This study aims to analyze and compare the performance of two artificial intelligence–based image restoration methods, namely GFPGAN (Generative Facial Prior GAN) and RestoreFormer, in enhancing the quality of old photographs of TGKH. Muhammad Zainuddin Abdul Madjid. Both methods were tested on images that had undergone visual degradation, with OpenCV used as a conventional baseline. The evaluation was conducted using two main parameters: processing time and result quality, measured through the Peak Signal-to-Noise Ratio (PSNR) metric. The results show that GFPGAN produced the best outcome, achieving the highest PSNR value (32.8 dB) and the fastest processing time (9 seconds), generating sharp and realistic facial details. RestoreFormer yielded nearly comparable quality with a PSNR of approximately 30 dB and a slightly longer processing time (10 seconds), but it demonstrated greater consistency in preserving the authenticity of textures and structural details. Meanwhile, the manual OpenCV-based method achieved only moderate improvement (25.2 dB) with the longest processing time (12.5 seconds). These findings indicate that AI-based technologies, particularly GFPGAN and RestoreFormer, hold great potential for the preservation of historical visual archives, as they can significantly enhance the quality of old photographs while maintaining their original visual character. The combination of both methods is recommended for digital restoration efforts that balance technical efficiency and cultural authenticity.

Keywords

Old Photo Restoration; GFPGAN; RestoreFormer; Artificial Intelligence; PSNR.

Full Text:

References

X. Chen, J. Tan, T. Wang, K. Zhang, W. Luo, and X. Cao, “Toward Real-World Blind Face Restoration With Generative Diffusion Prior,” IEEE Trans. Circuits Syst. Video Technol., vol. 34, no. 9, pp. 8494–8508, 2024, doi: 10.1109/TCSVT.2024.3383659.

M. Djamaluddin, M. N. Wathani, and H. Bahtiar, “Pelestarian Karya TGKH . Muhammad Zainuddin Abdul Madjid Melalui Digitalisasi Teks Dan Pengembangan Aplikasi Berbasis Mobile,” J. Pendidikan, Sains, Geol. dan Geofis., vol. 6, no. 1, 2025, [Online]. Available: https://jpfis.unram.ac.id/index.php/GeoScienceEdu/article/view/819.

H. Bahtiar, M. Djamaluddin, and L. M. Samsu, “Digitization of Hizib Nahdlatul Wathan by TGKH. Muhammad Zainuddin Abdul Madjid Based on the OPF Flipbook,” 2020.

H. Bahtiar, A. E. Sutriadi, M. Djamaluddin, and T. Barat, “Digitalisasi Kitab Nadham Batu Ngompal Karya TGKH . Muhammad Zainuddin Abdul Madjid Berbasis OPF Flipbook,” Infotek J. Inform. dan Teknol., vol. 3, no. 2, pp. 1–8, 2020.

X. Qiu, C. Han, Z. Zhang, B. Li, T. Guo, and X. Nie, “DiffBFR: Bootstrapping Diffusion Model for Blind Face Restoration,” MM 2023 - Proc. 31st ACM Int. Conf. Multimed., pp. 7785–7795, 2023, doi: 10.1145/3581783.3611731.

T. Yang, P. Ren, X. Xie, and L. Zhang, “GaN Prior Embedded Network for Blind Face Restoration in the Wild,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., pp. 672–681, 2021, doi: 10.1109/CVPR46437.2021.00073.

T. Wang et al., “A Survey of Deep Face Restoration: Denoise, Super-Resolution, Deblur, Artifact Removal,” ACM Comput. Surv., vol. 1, no. 1, pp. 1–35, 2022, [Online]. Available: http://arxiv.org/abs/2211.02831.

X. Lin et al., “DiffBIR: Toward Blind Image Restoration with Generative Diffusion Prior,” Lect. Notes Comput. Sci. (including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics) , vol. 15117 LNCS, pp. 430–448, 2025, doi: 10.1007/978-3-031-73202-7_25.

Z. Wang, J. Zhang, T. Chen, W. Wang, and P. Luo, “RestoreFormer++: Towards Real-World Blind Face Restoration From Undegraded Key-Value Pairs,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 45, no. 12, pp. 15462–15476, 2023, doi: 10.1109/TPAMI.2023.3315753.

S. W. Zamir, A. Arora, S. Khan, M. Hayat, F. S. Khan, and M. H. Yang, “Restormer: Efficient Transformer for High-Resolution Image Restoration,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2022-June, pp. 5718–5729, 2022, doi: 10.1109/CVPR52688.2022.00564.

M. Suin and R. Chellappa, “CLR-Face: Conditional Latent Refinement for Blind Face Restoration Using Score-Based Diffusion Models,” IJCAI Int. Jt. Conf. Artif. Intell., pp. 1290–1298, 2024.

Z. Wang et al., “Analysis and Benchmarking of Extending Blind Face Image Restoration to Videos,” IEEE Trans. Image Process., vol. 33, pp. 5676–5687, 2024, doi: 10.1109/TIP.2024.3463414.

O. Bouguerra, B. Attallah, and Y. Brik, “MRI-based brain tumor ensemble classification using two stage score level fusion and CNN models,” Egypt. Informatics J., vol. 28, no. October, p. 100565, 2024, doi: 10.1016/j.eij.2024.100565.

G. Azzopardi and A. Greco, “FaiResGAN : Fair and robust blind face restoration with biometrics preservation,” Image Vis. Comput., vol. 160, no. April, p. 105575, 2025, doi: 10.1016/j.imavis.2025.105575.

Hariman Bahtiar; Muhamad Wasil; Bambang Harianto, “Digitalisasi Karya TGKH. Muhammad Zainuddin Abdul Madjid berbasis mobile sebagai media pembelajaran interaktif untuk pelestarian peninggalan pendiri Nahdlatul Wathan,” Infotek J. Inform. dan Teknol., vol. 88, no. 5, pp. 427–434, 2019, doi: 10.2207/jjws.88.427.

X. Cheng, W. Siu, and L. Ieee, “Edge fusion back projection GAN for large scale face super resolution ☆,” J. Vis. Commun. Image Represent., vol. 100, no. August 2023, 2024, doi: 10.1016/j.jvcir.2024.104143.

S. S. Sciencedirect, “ScienceDirect ScienceDirect PSNR and SSIM : Evaluation of the Imperceptibility Quality of PSNR and SSIM : Evaluation of the Imperceptibility Quality of Images Transmitted over Wireless Networks Images Transmitted over Wireless Networks,” vol. 00, 2024, doi: 10.1016/j.procs.2024.11.134.

H. Zhao, Y. Gou, B. Li, D. Peng, J. Lv, and X. Peng, “Comprehensive and Delicate: An Efficient Transformer for Image Restoration,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2023-June, pp. 14122–14132, 2023, doi: 10.1109/CVPR52729.2023.01357.

N. Gao et al., “DiffMAC: Diffusion Manifold Hallucination Correction for High Generalization Blind Face Restoration,” 2024, [Online]. Available: http://arxiv.org/abs/2403.10098.

H. Yadav, “ScienceDirect Procedia ScienceDirect Image Restoration with Conditional Noise Diffusion : Restoring Image Restoration with Conditional Noise Diffusion : Restoring Facial Clarity and Details Facial Clarity and Details,” Procedia Comput. Sci., vol. 258, pp. 1159–1168, 2025, doi: 10.1016/j.procs.2025.04.350.

J. Sun, “ScienceDirect Enhancement and Optimization of Drone Oblique Photography Data Based on Generative Adversarial Network ( GAN ),” Procedia Comput. Sci., vol. 261, pp. 320–326, 2025, doi: 10.1016/j.procs.2025.04.210.

Z. Wang et al., “DR2: Diffusion-Based Robust Degradation Remover for Blind Face Restoration,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., vol. 2023-June, pp. 1704–1713, 2023, doi: 10.1109/CVPR52729.2023.00170.

G. Azzopardi and A. Greco, “FaiResGAN : Fair and robust blind face restoration with biometrics preservation,” Image Vis. Comput., vol. 160, no. October 2024, p. 105575, 2025, doi: 10.1016/j.imavis.2025.105575.

DOI: https://doi.org/10.30591/jpit.v11i2.10291