Skip to content
Licensed Unlicensed Requires Authentication Published by Oldenbourg Wissenschaftsverlag April 1, 2022

Image Garden

Curating Collections and Designing Smart Exhibitions with AI-Based Tools

Eugenia Sinatti, Simon Weckert and Ewelina Dobrzalski
From the journal i-com


Image Garden is an AI-based toolbox for curating collections and designing smart exhibitions that was developed by ART+COM Studios [ART+COM Studios. 2021. ART+COM.] in the research project QURATOR [QURATOR Bündnis. 2021. Qurator.]. This paper describes the application scenarios underlying the development of Image Garden. It also presents the technologies integrated in its processing pipeline, and sketches Show Cases that were implemented to demonstrate and evaluate functionality and usability.

Funding source: Bundesministerium für Bildung und Forschung

Award Identifier / Grant number: 03WKDA1D

Funding statement: The work presented in this paper was partially supported by the Federal Ministry of Education and Research (BMBF) in the context of the research project QURATOR (03WKDA1D).


The success and final outcome of this paper would not have been possible without the close collaboration with our dear colleagues from ART+COM Studios. We are especially grateful to the advice and guidance of Dr. Joachim Quantz, his trust and encouragement.

We also want to thank Dr. Joachim Böttger, for his technical support and fruitful ideas that he shared with us while doing the research and Prof. Jussi Ängeslevä, for his support especially in the important first moment of the project. A big thank you goes also to the Bauhaus-Archiv that gave us access to their database of their collection. Without all this help we would not have been able to realize this publication and this work.


[1] Alexandr Andoni, Piotr Indyk, and Ilya Razenshteyn. 2018. Approximate Nearest Neighbor Search in High Dimensions. arXiv:cs.DS/1806.09823.Search in Google Scholar

[2] 10625 Berlin Bauhaus-Archiv e. V. | Museum für Gestaltung, Schillerstraße 9. 2021. Bauhaus-Archiv. in Google Scholar

[3] Internationale Beleuchtungskommision. [n. d.]. CIELAB. in Google Scholar

[4] Erik Bernhardsson. 2021. Ann benchmarks. in Google Scholar

[5] QURATOR Bündnis. 2021. Qurator. in Google Scholar

[6] The SciPy community. 2021. K-Dimensional Tree. in Google Scholar

[7] 10559 Berlin DFKI GMBH, Alt-Moabit 91C. 2021. Digitale Kutarierung. in Google Scholar

[8] Piotr Dollaŕ and C. Lawrence Zitnick. [n. d.]. Structured Forests for Fast Edge Detection. 2013 ([n. d.]).10.1109/ICCV.2013.231Search in Google Scholar

[9] Python Software Foundation. 2021. Annoy. in Google Scholar

[10] Python Software Foundation. 2021. Ridge Detection. in Google Scholar

[11] Bundesministerium für Bildung und Forschung. 2021. BMBF. in Google Scholar

[12] Google. 2021. “Draw to Art” by Google Creative Lab, Google Arts & Culture Lab. in Google Scholar

[13] Google. 2021. “Quick, Draw!” by Google Arts & Culture Lab. in Google Scholar

[14] Jing He and Joachim Quantz. [n. d.]. Interactive knowledge visualization tools for exhibition curation. International Journal of Computing, Article 3 ([n. d.]), 173–170 pages.Search in Google Scholar

[15] Keras. 2021. Keras. in Google Scholar

[16] Scikit learn developers. 2020. KMeans. in Google Scholar

[17] McInnes Leland, Healy John, and Melville James. 2018. UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction. in Google Scholar

[18] museum digital. 2021. SMB, Staatliche Museen zu Berlin. in Google Scholar

[19] Nanonets. 2021. Data Augmentation | How to use Deep Learning when you have Limited Data Part 2. in Google Scholar

[20] Inc. Observable. 2021. Fashion MNIST. in Google Scholar

[21] François Pirsch. 2015. Vantage Point Trees. in Google Scholar

[22] Weigandufer 45 12059 Berlin pixolution GmbH, c/o Büro 2.0. 2021. Pixolution. in Google Scholar

[23] NVIDIA Research Projects. 2021. StyleGAN2-Ada Github Repository. in Google Scholar

[24] NVIDIA Research Projects. 2021. StyleGAN2 Github Repository. in Google Scholar

[25] Leland McInnes Revision. 2018. UMAP. in Google Scholar

[26] SmugMug+Flickr. 2021. flickr. in Google Scholar

[27] Princeton University Stanford Vision Lab, Stanford University. 2020. ImageNet. in Google Scholar

[28] ART+COM Studios. 2021. ART+COM. in Google Scholar

[29] ART*COM Studios. 2021. NuForm. in Google Scholar

[30] Christian Szegedy, Wei Liu, Yangqing Jia, Sermanet Pierre, Reed Scott, Anguelov Dragomir, Erhan Dumitru, Vincent Vanhoucke, and Rabinovich Andrew. 2015. Going deeper with convolutions. In 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 1–9. in Google Scholar

[31] Tensorflow. 2021. Tensorflow. in Google Scholar

[32] Karras Tero, Laine Samuli, and Aila Timo. 2019. A Style-Based Generator Architecture for Generative Adversarial Networks. In 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 4396–4405. in Google Scholar

[33] Alexander von Humboldt. 2017. Das graphische Gesamtwerk. Lambert Schneider Verlag. 800 pages.Search in Google Scholar

[34] Ross Wightman. 2016. ImageNet Script. in Google Scholar

[35] Digital Humanities Laboratory Yale University Library. [n. d.]. PixPlot. in Google Scholar

[36] Digital Humanities Laboratory Yale University Library. 2021. Yale Digital Humanities Lab Team. in Google Scholar

Published Online: 2022-04-01
Published in Print: 2022-04-26

© 2022 Walter de Gruyter GmbH, Berlin/Boston

Scroll Up Arrow