From Pampas to Pixels: Fine-Tuning Diffusion Models for Ga\'ucho Heritage

Generative AI has become pervasive in society, witnessing significantadvancements in various domains. Particularly in the realm of Text-to-Image(TTI) models, Latent Diffusion Models (LDMs), showcase remarkable capabilitiesin generating visual content based on textual prompts. This paper addresses thepotential of LDMs in representing local cultural concepts, historical figures,and endangered species. In this study, we use the cultural heritage of RioGrande do Sul (RS), Brazil, as an illustrative case. Our objective is tocontribute to the broader understanding of how generative models can help tocapture and preserve the cultural and historical identity of regions. The paperoutlines the methodology, including subject selection, dataset creation, andthe fine-tuning process. The results showcase the images generated, alongsidethe challenges and feasibility of each concept. In conclusion, this work showsthe power of these models to represent and preserve unique aspects of diverseregions and communities.