Automatic Analysis of Human Body Representations in Western Art

The way the human body is depicted in classical and modern paintings relevant for art historical analyses. Each artist has certain themes concerns, resulting different poses being used more heavily than others. In this paper, we propose a computer vision pipeline to analyse pose representations paintings, which can be specific artists or periods. Specifically, combine two estimation approaches (OpenPose DensePose, respectively) introduce methods deal with occlusion perspective issues. For normalisation, map detected contours Leonardo da Vinci’s Vitruvian Man, depiction of proportions. We visualisation approach illustrating articulation joints set paintings. Combined hierarchical clustering poses, our reveals common uncommon by artists. Our improves over purely skeleton based analyses