Searching for cultural relationships through deep learning models

Family album photo collections may reveal historical insights regarding specific cultures and times. In most cases, such photos are scattered among private homes and only available on paper or photographic film, thus making their analysis very cumbersome. Their study may also become difficult because of the number of photos that such collections contain. It would be exceedingly long to manually verify the characteristics of more than a few hundred photos, considering that often no associated descriptions are available. This work falls in the described domain, addressing the problem of dating an image resorting to the analysis of an analog family album photo dataset, namely IMAGO, containing photos shot in the 20th century. Thanks to the IMAGO dataset, it was possible to apply different deep learning-based architectures to date images belonging to photo albums without needing any other sources of information. In addition, with the implementation of cross-dataset experiments, which also involved models previously presented in the literature, it was possible to observe temporal shifts which may be due to known intercultural influences. Concluding, deep learning models revealed their potential not only in terms of their performance but also in terms of their possible applications to intercultural research.