KIHT: Kaligo-based Intelligent Handwriting Teacher
KIHT project overview
The KIHT project has the objective of creating an intelligent learning tool known as the "DigiPen," which is an electronic pen designed for digitized handwriting. STABILO, in collaboration with the German KIT institute, is responsible for developing the hardware of the electronic DigiPen. The KIT institute is focused on integrating artificial intelligence algorithms into the device. The IRISA IntuiDoc team is in charge of designing and developing the deep learning AI system that can reconstruct online handwriting patterns from the data captured by the DigiPen's kinematic sensors, including accelerometers, gyroscopes, magnetometers, and force sensors.
My contribution as an IRISA member
My work has focused mainly on the pre-processing chain for processing, synchronizing and labeling data (sensor and trajectory) to make it usable for deep neural network learning. We collected a significant amount of data using the latest generation Digipen. The pre-processing chain has been finalized. It is based on automatic alignment between ground truth (trajectory captured on digital tablet) and DigiPen sensor data (kinematic sensor) using the Dynamic Time Warping (DTW) algorithm. A neural network architecture based on Temporal Convolutional Networks (TCN) was designed and its hyperparameters optimized, and an evaluation based on Fréchet distance was implemented for optimum analysis of the results.
Related papers
Article in an international peer-reviewed journal (+ presentation at ICDAR 2023 – Journal Track)
Communication at the SIFED symposium in France without proceeding
Conference poster at the SIFED symposium in France without proceeding
Technical Paper)