Text2Doppler Dataset

This is a demo dataset generated by the Text2Doppler pipeline [1]. The data includes the skeleton data generated from motion descriptions and the corresponding synthetic micro-Doppler data set. The variations in viewing angle, motion speed and background are not included. The dataset can be used to study the effect of data diversity in radar HAR tasks. If you use it in your work, please cite the corresponding paper. Also, if you need to generate your own data and introduce more types of diversity, you can email me. [1] Text2Doppler: Generating Radar Micro–Doppler Signatures for Human Activity Recognition via Textual Descriptions. https://ieeexplore.ieee.org/document/10670276