We introduce an interdisciplinary project for archaeological and computer vision research teams on the analysis of the ancient Maya writing system. Our first task is the automatic retrieval of Maya syllabic glyphs using the Shape Context descriptor. We investigated the effect of several parameters to adapt the shape descriptor given the high complexity of the shapes and their diversity in our data. We propose an improvement in the cost function used to compute similarity between shapes making it more restrictive and precise. Our results are promising, they are analyzed via standard image retrieval measurements.