Visualization-Driven Illumination for Density Plots

An interactive prototype for the technique proposed in our IEEE TVCG submission "Visualization-Driven Illumination for Density Plots".

The paper introduces an innovative density plot design that integrates two key components: (i) a visualization-driven illumination model tailored to enhance density-plot-specific analysis tasks, and (ii) a novel image composition technique to reduce the interference between image shading and color-encoded density values. VIDP is able to maintain the original colors and reveal low-density outliers at the same time, and thus provides a more accurate perception of relative data density than existing techniques.

Demo

You can try our technique through the online application available at https://xinchen-sdu.github.io/Visualization-Aware-Illumination-for-Density-Plots/.

The format of the input custom dataset should be a CSV file consisting of two columns of numbers without a header.

Project Setup

We offer two versions to experience the VIDP technique, you can choose whichever you prefer.

JavaScript version

We developed this prototype on Ubuntu 20.04 LTS with Vue 3, Element Plus, D3.js, and fast-kde.

1. Clone this repository

2. Install necessary dependencies

npm install

3. Compile and Hot-Reload for Development

npm run dev

4. Compile and Minify for Production

npm run build

BTW, you can lint the project with ESLint

npm run lint

Python version

We tested our code on Ubuntu 20.04 LTS with Python 3.8.10.

1. Clone this repository

2. Install the following dependencies

pip install notebook==6.5.6 numpy==1.21.5 pandas==1.4.3 matplotlib==3.6.3 opencv-contrib-python==4.10.0.82 KDEpy==1.1.8 scikit-learn==1.1.3 pycolormap-2d==1.0.10

3. Start jupyter notebook

jupyter notebook

4. Open illumination.ipynb

License

This code is licensed under the MIT License. If you use our data or code in your work, please cite our paper:

@article{chen2025vidp,
  author={Chen, Xin and Wang, Yunhai and Bao, Huaiwei and Lu, Kecheng and Jo, Jaemin and Fu, Chi-Wing and Fekete, Jean-Daniel},
  journal={IEEE Transactions on Visualization and Computer Graphics}, 
  title={Visualization-Driven Illumination for Density Plots}, 
  year={2025},
  volume={31},
  number={2},
  pages={1631-1644},
  keywords={Image color analysis;Lighting;Data visualization;Visualization;Distortion;Kernel;Estimation;Computational modeling;Analytical models;Shape;Density plot;illumination;image composition;shading},
  doi={10.1109/TVCG.2024.3495695}
}