Construcción de una representación basada en análisis formales de obras de arte, aproximando aspectos de la apreciación artística, que eventualmente pueda ser utilizada en la generación y evaluación de artefactos visuales

Discovering influences between paintings and artists is very important for automatic art analysis. Lately this problem has gained more importance since researches are looking into explanations about origin and evolution of artistic styles. This research proposes a methodology to build an image representation based on formal analyzes and design principles present in art works, trying to approximate as much as possible artistic appreciation characterized on the ground of three theoretical models. Our approach starts with a hierarchical image segmentation process. Based on the result of this first step, we build a straight skeleton over every segment. Then we extract some color information of every region and suggest a color palette that clusters 180 samples. Afterwards a shape analysis and classification procedure are implemented. Then some design principles are calculated over the complete graph induced by the centroids of every main region of every art work. Finally, we build the vector-form of the proposed representation. With the generated representation, we processed to construct a distance hierarchy based on the previous shape classification. Using this distance hierarchy, we define a similarity function that takes into account the categorical and numerical dimensions of our representation. We apply the GHSOM procedure to several aspects of our representation to be able to generate several complemented new versions of networks based on the previous notion of Creativity Implication Network. We used the recently proposed MultiRank algorithm to suggest possible ways to give interesting insights into the objectives of this research. Our results corroborate some well-known facts about the artists analyzed. We plan to expand our analysis to include more abstract artworks and also suggest some future work related to computational creative systems, that might let us validate more our results and test how our methodology could be used to generate visual artifacts too.

• Autores: Gutiérrez García, Luis Fernando
• Tipo de recurso: Doctoral thesis
• Fecha de publicación: 2020
• Institución: Universidad de los Andes
• Repositorio: Séneca: repositorio Uniandes
• Idioma: spa
• OAI Identifier: oai:repositorio.uniandes.edu.co:1992/50833
• Acceso en línea: http://hdl.handle.net/1992/50833
• Palabra clave: Metodología en apreciación del arte; Creatividad en el arte; Cartografía en el arte; Crítica de arte; Ingeniería
• Rights: openAccess
• License: https://repositorio.uniandes.edu.co/static/pdf/aceptacion_uso_es.pdf