Direct visualization techniques for the analysis of image data: the slice histogram and the growing entourage plot
AbstractThe following is a description of two novel techniques for the direct visualization of image data. Direct visualizations of image data make use of the images in their original visible format. The first technique, the slice histogram, arranges slices of images as histograms, organized by both visual and non-visual variables. The second technique, the growing entourage plot, organizes high-dimensional clusters of images on a 2D canvas by projection. Both techniques are designed for exploratory analysis of image datasets.
Manovich, L. (2011). What is visualization? Visual Studies, Vol. 26, no. 1, 36-49. Available at http://manovich.net/index.php/projects/what-is-visualization.
Manovich, L. (2011). Media visualization: Visual techniques for exploring large media collections. The International Encyclopedia of Media Studies Vol. VI: Media Studies Futures, edited by Kelly Gates (Chichester: Wiley Blackwell). Available at http://manovich.net/index.php/projects/media-visualization-visual-techniques-for-exploring-large-media-collections.
Smeulders, A. W., Worring, M., Santini, S., Gupta, A., & Jain, R. (2000). Content-based image retrieval at the end of the early years. IEEE Transactions on Transactions on Pattern Analysis and Machine Intelligence, 22(12), 1349-1380.
Van der Maaten, L., & Hinton, G. (2008). Visualizing data using t-SNE. Journal of Machine Learning Research, 9 (2579-2605), 85.
Yazdani, M., & Manovich, L. (2015). Predicting social trends from non-photographic images on Twitter, 2015 IEEE International Conference on Big Data, 1653-1660. Available at http://manovich.net/index.php/projects/predicting-social-trends