Find visual trends in newspaper advertisements with SIAMESE


Over the past few months, Melvin Wevers joined the KB Lab as Researcher-in-residence. During his stay, he did not only manage to defend his PhD succesfully, but together with Juliette Lonij, he worked on a research project focusing on finding trends in newspaper advertisements using computer vision methods.

This work has now been concluded and a wonderful advertisement browsing tool using an underlying dataset of almost 430.000 advertisements are available for research. Melvin describes the process of the project and how to use the tool in his blog post. We would thus like to invite you to have a look at the blog, try the tool and let us know if you would like to also use the data.

Blog: Find What You Were Looking For: Using Neural Networks to Trawl through Newspaper Advertisements

Advertisements offer us a sneak peek into the ideals and aspirations of past realities. They show the state of technology, the social functions of products, and provide information on the society in which a product was to be sold. Continue reading...


Delpher contains millions of advertisements in digital form. However, querying for advertisements with a specific visual style or ads for a particular product is more cumbersome than one would expect. Continue reading and try the tool...


The SIAMESET dataset was created by Melvin Wevers and Juliette Lonij for the tool SIAMESE. The dataset consists of images and metadata of advertisements from two Dutch national newspapers: Algemeen Handelsblad (1945-1969) and NRC Handelsblad (1970-1994). Continue reading and access the set...