We present a non-uniform recursive sampling technique for multi-class scatterplots, with the specific goal of faithfully presenting relative data and class densities, while preserving major outliers in the plots. Our technique is based on a customized binary kd-tree, in which leaf nodes are created by recursively subdividing the underlying multi-class density map. By backtracking, we merge leaf nodes until they encompass points of all classes for our subsequently applied outlier-aware multi-class sampling strategy. A quantitative evaluation shows that our approach can better preserve outliers and at the same time relative densities in multi-class scatterplots compared to the previous approaches, several case studies demonstrate the effectiveness of our approach in exploring complex and real world data.
@article{Chen2019RecursiveSubdivisionTechnique, author = {X. Chen, T. Ge, J. Zhang, B. Chen, C. Fu, O. Deussen, Y. Wang}, doi = {10.1109/TVCG.2019.2934541}, issn = {2160-9306}, journal = {IEEE Transactions on Visualization and Computer Graphics}, keywords = {Visualization;Data visualization;Measurement;Sampling methods;Estimation;Clutter;Image color analysis;Scatterplot;multi-class sampling;kd-tree;outlier;relative density}, number = {1}, pages = {729--738}, title = {A Recursive Subdivision Technique for Sampling Multi-class Scatterplots}, url = {http://graphics.uni-konstanz.de/publikationen/Chen2019RecursiveSubdivisionTechnique}, volume = {26}, year = {2020} }