Assessing a Shape Descriptor for Analysis of Mesoamerican Hieroglyphics: A View Towards Practice in Digital Humanities

Technological advances in digitization, automatic image analysis and information management are enabling the possibility to analyze, organize and visualize large cultural datasets. As one of the key visual cues, shape feature has been used in various image analysis tasks such as handwritten character recognition [1, 5], sketch analysis [4], etc. We assess a shape descriptor, within the application domain of Maya hieroglyphic analysis. Our aim is to introduce this descriptor to the wider Digital Humanities (DH) community, as a shape analysis tool for DH-related applications. The Maya civilization is one of the major cultural developments in ancient Mesoamerica. The ancient Maya language infused art with uniquely pictorial forms of hieroglyphic writing, which represents an exceptionally rich legacy [10]. Most Maya texts were written during the Classic period (AD 250-900) of the Maya civilization on various media types, including stone monuments. A special class of Maya texts was written on bark cloths as folding books from the Post-Classic period (AD 1000-1519). Only three such books (namely the Dresden, Madrid and Paris codices) are known to have survived the Spanish Conquest. A typical Maya codex page contains icons, main sign glyph blocks, captions, calendric signs, etc. Fig. 1 illustrates an example page segmented into main elements [6]. In this paper, we are interested in the main signs. Maya hieroglyphic analysis requires epigraphers to spend a significant amount of time browsing