»The Curator‘s Machine« is a tool that assists curators by utilizing either OpenAI’s CLIP model or the OpenCLIP model and provides an AI-guided search that is able to:
… adapt the search results to the current session to suggest images that fit the project at hand
… search your own image collection
… search with an intuitive and user-defined description (e.g., a dog in front of a tree)
… search by image to get similar images
… filter the search based on Metadata (if provided)
»The Curator‘s Machine« was developed in cooperation with the Visual Computing Institute at RWTH Aachen University.
Developers: Moritz Ibing, Isaak Lim, Marian Schneider, Tim Elsner and Leif Kobbelt.
Concept: Dominik Bönisch.
Tests and Feedback: Nora Riediger, Sonja Benzner, Eva Birkenstock, Ana Salazar, Lisa Oord, Esther Boehle and Galina Dekova.
The GitHub Repository can be found here: https://github.com/VCI-RWTH/TrainingTheArchive