2025-08-202025-08-202025-07-17FERREIRA, Marcus Vinícius Carvalho. Arte acessível: uma análise comparativa de algoritmos de detecção de obras de arte. Orientador: Iago Lins de Medeiros. 2025. 27 f. Trabalho de Conclusão de Curso (Bacharelado em Engenharia da Computação) – Faculdade de Engenharia da Computação, Campus Universitário de Tucuruí, Universidade Federal do Pará, Tucuruí, 2025. Disponível em: https://bdm.ufpa.br/handle/prefix/8496. Acesso em:.https://bdm.ufpa.br/handle/prefix/8496Computer vision has proven to be a promising tool in the development of assistive technologies, particularly in promoting accessibility for people with visual impairments. This work presents a comparative analysis of object detection algorithms focused on the recognition of painted artworks, using convolutional neural networks. The evaluated models include YOLOv8, YOLOv11, RetinaNet, and Faster R-CNN, with emphasis on their effectiveness and feasibility for deployment on mobile devices. The dataset was created through the automated collection of images from the internet, featuring renowned artworks such as Mona Lisa, Meisje met de Parel, and The Starry Night. All images were manually annotated using the LabelImg tool. The models were trained and assessed based on metrics such as mean Average Precision (mAP), recall, inference time, and computational resource usage. The results showed that the YOLO-based models, especially YOLOv8, achieved the best balance between performance and efficiency, reaching a mAP of 0.992 and recall of 0.987. Additionally, they demonstrated shorter inference times, making them particularly suitable for real-time applications and devices with limited hardware capabilities. This research contributes to the advancement of accessible solutions in the fields of art and digital inclusion, highlighting the potential of artificial intelligence as an ally in promoting cultural accessibility.Acesso AbertoVisão computacionalAcessibilidadeDetecção de objetosComputer visionAccessibilityObject detectionCNPQ::ENGENHARIASCNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::METODOLOGIA E TECNICAS DA COMPUTACAOCNPQ::LINGUISTICA, LETRAS E ARTES::ARTES::ARTES PLASTICAS::PINTURATrabalho de Curso - Graduação - MonografiaAttribution-NonCommercial-NoDerivs 3.0 Brazil