Exploring the Intersection of Technology and History: Stable Diffusion Inpainting in Heritage Restoration
Kajol Chaudhary
Student of MA in Modern History, Sharda School of Humanities & Social Sciences, Sharda University, India.
Tanni Moitra
Assistant Professor of History, Sharda School of Humanities & Social Sciences, Sharda University, India.
Preserving cultural heritage is essential for sustaining historical knowledge and architectural continuity. This study examines the use of Stable Diffusion Inpainting for the digital restoration of partially damaged monuments, with focus on the gopurams of the Vittala Temple in Hampi and Salimgarh Fort in Delhi. The research applies the Runway ML v1.5 Stable Diffusion Inpainting model to reconstruct missing architectural elements preserving spatial accuracy and structural balance. The methodology involves image preprocessing through noise reduction and damaged region identification, followed by systematic model implementation and evaluation. The findings show that the model effectively restores complex forms and ornamental details with high visual coherence and minimal structural distortion. The case studies confirm practical applicability also identifying limitations related to texture alignment and historical precision. Overall, the study finds that Stable Diffusion Inpainting is a reliable digital tool that can support heritage conservation and informed restoration practices.
Keywords: Stable Diffusion, Inpainting, Heritage Restoration, Machine Learning, and Digital Preservation.
Ahmed, H. O., Alfaqheri, T., & Sadka, A. H. (2021). Digital image inpainting techniques for cultural heritage preservation and restoration. In J. Smith & R. Johnson (Eds.), Data analytics for cultural heritage: Current trends and concepts (pp. 91–122). Springer.Barnes, C., Shechtman, E., Finkelstein, A., & Goldman, D. B. (2009). PatchMatch: A randomized correspondence algorithm for structural image editing. ACM Transactions on Graphics, 28(3), 24.Bertalmio, M., Sapiro, G., Caselles, V., & Ballester, C. (2000). Image inpainting. In Proceedings of the 27th Annual Conference on Computer Graphics and Interactive Techniques (pp. 417–424). ACM.Childers, D. G. (1979). Digital restoration from start to finish: How to repair old and damaged photographs. W. W. Norton & Company.Efros, A. A., & Leung, T. K. (1999). Texture synthesis by non-parametric sampling. In Proceedings of the Seventh IEEE International Conference on Computer Vision (Vol. 2, pp. 1033–1038). IEEE.Fritz, J. M., Michell, G., & Nagaraja Rao, M. S. (1984). Where kings and gods meet: The royal centre at Vijayanagara, India. University of Arizona Press.Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., & Bengio, Y. (2014). Generative adversarial nets. In Advances in Neural Information Processing Systems (pp. 2672–2680). Curran Associates.Jin, M., Yang, J., & Wu, E. (2017). Stable and fast high-dimensional Gaussian filtering for image inpainting. IEEE Transactions on Image Processing, 26(8), 3781–3794.Liu, H., Xu, Q., & Wu, S. (2021). Stable and efficient inpainting via diffusion and iterative thresholding. IEEE Transactions on Image Processing, 30, 1627–1642.McCullough, M. (1998). Abstracting craft: The practiced digital hand. MIT Press.Mendoza, M. A. D., De La Hoz Franco, E., & Gómez, J. E. G. (2023). Technologies for the preservation of cultural heritage: A systematic review of the literature. Sustainability, 15(2), 1059.Michell, G. (1989). The sacred centre as a symbol of political power. In Vijayanagara: Architectural inventory of the sacred centre (pp. 37–53). American Institute of Indian Studies.Nath, R. (2005). History of Mughal architecture: Vol. IV (The age of Shah Jahan, Part II). Abhinav Publications.Pathak, D., Krähenbühl, P., Donahue, J., Darrell, T., & Efros, A. A. (2016). Context encoders: Feature learning by inpainting. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 2536–2544). IEEE.Rombach, R., Blattmann, A., Lorenz, D., Esser, P., & Ommer, B. (2022). High-resolution image synthesis with latent diffusion models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 10684–10695). IEEE.Wagner, A., & de Clippele, M. S. (2023). Safeguarding cultural heritage in the digital era–A critical challenge. International Journal for the Semiotics of Law-Revue internationale de Sémiotique juridique, 36(5), 1915–1923.Wang, Z., Bovik, A. C., Sheikh, H. R., & Simoncelli, E. P. (2004). Image quality assessment: From error visibility to structural similarity. IEEE Transactions on Image Processing, 13(4), 600–612Yu, J., Lin, Z., Yang, J., Shen, X., Lu, X., & Huang, T. S. (2018). Generative image inpainting with contextual attention. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 5505–5514). IEEEZhang, R., Isola, P., & Efros, A. A. (2018). The unreasonable effectiveness of deep features as a perceptual metric. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 586–595). IEEE.Zhang, Y., Yang, Y., & Wang, H. (2019). Fast image inpainting with Fourier method. Journal of Visual Communication and Image Representation, 59, 87–95.Zhang, X. (2024). AI-Assisted Restoration of Yangshao Painted Pottery Using LoRA and Stable Diffusion. Heritage (2571-9408), 7(11).
