Ancient Roman Coin Recognition in the Wild Using Deep Learning Based Recognition of Artistically Depicted Face Profiles

Abstract

As a particularly interesting application in the realm of cultural heritage on the one hand, and a technically challenging problem, computer vision based analysis of Roman Imperial coins has been attracting an increasing amount of research. In this paper we make several important contributions. Firstly, we address a key limitation of existing work which is largely characterized by the application of generic object recognition techniques and the lack of use of domain knowledge. In contrast, our work approaches coin recognition in much the same way as a human expert would: by identifying the emperor universally shown on the obverse. To this end we develop a deep convolutional network, carefully crafted for what is effectively a specific instance of profile face recognition. No less importantly, we also address a major methodological flaw of previous research which is, as we explain in detail, insufficiently systematic and rigorous, and mired with confounding factors. Lastly, we introduce three carefully collected and annotated data sets, and using these demonstrate the effectiveness of the proposed approach which is shown to exceed the performance of the state of the art by approximately an order of magnitude.

Understanding Ancient Coin Images

“…Deep learning has proven successful in a range of image understanding tasks [16]. Recent pioneering work on its use on images of coins has demonstrated extremely promising results, outperforming more traditional approaches by an order of magnitude [15]. A major difference in the nature of the challenge we tackle here is that our data set is weakly supervised -that is, samples are labelled only at the image level by a corresponding largely unstructured textual description, even though the actual semantic elements of interest themselves occupy a relative small area of the image.…”

Strategies for informed sample size reduction in adaptive controlled clinical trials

“…In particular, they make use of concepts such as heritable characteristics and fitness-based selection and have been applied with success in a number of diverse domains [15][16][17].…”

“…Equation 17 provides a ready means of assessing the relative fitness of two solutions-a chromosome encoding a solution associated with lesser loss is fitter than the one associated with a greater loss. However we still need a way of accounting for this in the random selection of chromosomes which mate to generate offspring.…”

“…(17). Specifically, we generate a hypothesis in the form of a set of samples considered for removal one by one.…”

“…After n r iterations the entire hypothesis has been generated and its quality can be assessed by the associated information loss given by Eq. (17). If n h hypotheses (i.e.…”