In this paper, we present a development for recognizing objects from looted excavations. Experts with required expertise are not always available where an archaeological object needs to be assessed for import, export or trade. For this purpose, we developed a smartphone app that can provide on-site assistance in the initial assessment of archaeological objects. The app sends captured images to a server for recognition and receives results with similar objects and their metadata along with an associated probability. A user can thus use these information to infer the provenance of the photographed object. To this end, a deep learning based solution was developed to identify archaeological objects, including a classifier trained using transfer learning and an image matching scheme based on deep convolutional neural networks (CNN) features. The developed application will be tested by law enforcement agencies with a total of 15 smartphones for six months starting in early October.

Identifying cultural assets is a challenging task which requires specific expertise. In this paper, a deep learning based solution to identify archaeological objects is proposed. Several additions to the ResNet CNN architecture are introduced which consolidate features from different intermediate layers by applying global pooling operations. Unlike general object recognition, identifying archaeological objects poses new challenges. To meet the special requirements in classifying antiques, a hybrid network architecture is used to learn the characteristics of objects using transfer learning, which includes a classification network and a regression network. With the help of the regression network, the age of objects can be predicted, which improves the overall performance in comparison to manually classifying the age of objects. The proposed scheme is evaluated using a public database of cultural assets and the experimental results demonstrate its significant performance in identifying antique objects.