Abstract: This paper presents a generative data augmentation approach for human activity recognition (HAR) to close the distribution gap between laboratory training and real-world deployment. Despite ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results