In the context of developing a pedagogical tool for teaching anatomy, the need for a comparative study between two RGB-D cameras has emerged. This paper addresses the assessment of the accuracy and precision of two RGB-D sensors (Carmine Primesense and Persee Orbbec) through two different experiments. The evaluation not only provides comparative results on the sensors performances but also aims at determining in which conditions they are the most efficient. The first experiment evaluates the variability of the output depth map data. The second experiment focuses on analyzing the influence of the distance in the positioning accuracy of an object submitted to controlled displacements. The results are summarized in a set of error heat maps and a table; they provide clues for using one sensor rather than another by describing their robustness both in a static scene and in a motion capture scenario.