Filtering
CSN allows users to filter and explore their collections using interactive and customizable Dimension Filters based on metadata and vector embeddings.
We introduce the Collection Space Navigator (CSN), a browser-based visualization tool to explore, research, and curate large collections of visual digital artifacts that are associated with multidimensional data, such as vector embeddings or tables of metadata. Media objects such as images are often encoded as numerical vectors, for e.g. based on metadata or using machine learning to embed image information. Yet, while such procedures are widespread for a range of applications, it remains a challenge to explore, analyze, and understand the resulting multidimensional spaces in a more comprehensive manner. Dimensionality reduction techniques such as t-SNE or UMAP often serve to project high-dimensional data into low dimensional visualizations, yet require interpretation themselves as the remaining dimensions are typically abstract. Here, the Collection Space Navigator provides a customizable interface that combines two-dimensional projections with a set of configurable multidimensional filters. As a result, the user is able to view and investigate collections, by zooming and scaling, by transforming between projections, by filtering dimensions via range sliders, and advanced text filters. Insights that are gained during the interaction can be fed back into the original data via ad hoc exports of filtered metadata and projections. This paper comes with a functional showcase demo using a large digitized collection of classical Western art. The Collection Space Navigator is open source. Users can reconfigure the interface to fit their own data and research needs, including projections and filter controls. The CSN is ready to serve a broad community.
The Collection Space Navigator (CSN) is an explorative visualization tool for researching collections and their multidimensional representations. We designed this tool to better understand multidimensional data, its methods, and semantic qualities through spatial navigation and filtering. CSN can be used with any image collection and can be customized for specific research needs (see Jupyter Notebook or Google Colab).
CSN allows users to filter and explore their collections using interactive and customizable Dimension Filters based on metadata and vector embeddings.
CSN provides customizable projections for visualizing and analyzing collections in two-dimensional space with the ability to zoom, scale, and switch between different projections.
CSN allows categorical data to be displayed using colored frames in the Projections Area to reveal patterns and clusters.
CSN can also be used to display 2D plots with other types of data, such as time series.
CSN supports complex and nested search queries with Boolean operators like AND, OR, equal (==) and not equal (!=).
CSN allows to export the metadata of the currently filtered objects as CSV file and the current projection view as PNG.
Using Jupyter Notebook or Google Colab, preparing datasets to build a customized version of the Collection Space Navigator is made easy.
@inproceedings{ohm2023collection,
title={Collection space navigator: an interactive visualization interface for multidimensional datasets},
author={Ohm, Tillmann and Canet Sola, Mar and Karjus, Andres and Schich, Maximilian},
booktitle={Proceedings of the 16th International Symposium on Visual Information Communication and Interaction},
pages={1--5},
year={2023}
}@misc{ohm2023collection,
title={Collection Space Navigator: An Interactive Visualization Interface for Multidimensional Datasets},
author={Tillmann Ohm
and Mar Canet Solà
and Andres Karjus
and Maximilian Schich},
year={2023},
eprint={2305.06809},
archivePrefix={arXiv},
primaryClass={cs.CV}
}Compression ensembles quantify aesthetic complexity and the evolution of visual art by Andres Karjus, Mar Canet Solà, Tillmann Ohm, Sebastian E. Ahnert, Maximilian Schich (2022)
The CSN code is partly based on the umap-explorer by GrantCuster.