We are happy to announce that we have just published a paper in “Frontiers in Artificial Intelligence” about our last research in the project OrthoKI.
📄 Title: A diffusion model for inertial based time series generation on scarce data availability to improve human activity recognition
🔗 Read it here: https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2025.1640948/full
In this paper, we introduce a new method to significantly accelerate the computation time for training diffusion models (DDPM). These models can be used to synthesize training data for AI models. In the paper, we show that this method can achieve a 20% to 60% acceleration in computation time without compromising data quality.
Huge thanks to our collaborators and team members for their contributions!
#AI #SensorData #MachineLearning #Publication