We’re excited to share our new paper we have just published in Nature Scientific Reports on synthetic human activity data generation using denoising diffusion models.
📄 Title: A diffusion model for inertial based time series generation on scarce data availability to improve human activity recognition
🔗 Read it here: https://www.nature.com/articles/s41598-025-01614-x
Our paper introduces a novel approach using denoising diffusion probabilistic models to synthetically generate multi-IMU data. This leads to a huge performance increase when training artificial neural networks on human activitiy recognition tasks.
Huge thanks to our collaborators and team members for their contributions!
#AI #SensorData #MachineLearning #Publication