Search papers, labs, and topics across Lattice.
This study introduces a Mixture-of-Experts (MOE) approach for reconstructing handwriting trajectories using a digital pen equipped with IMU sensors, focusing on both the touch and hover phases of writing. By training separate models for the touching and hovering phases, the method enhances trajectory accuracy and positioning for digital handwriting on any surface. The proposed approach outperforms existing methods, and the authors also present a novel public benchmark dataset to facilitate future research in handwriting reconstruction.
Achieving significant improvements in handwriting trajectory reconstruction, this study reveals that separating touch and hover dynamics can drastically enhance accuracy in digital pen applications.
The use of digital pens for online handwriting trajectory reconstruction is a prevalent method for human-computer interaction. In this study, we focus on a digital pen equipped with sensors where we aim at reconstructing the online handwriting trajectory. This pen enables writing on any surface and preserving the digital trace of handwriting. This type of pen could be used as an aid to learning to write in classroom. In this paper, we propose a new approach learning to finely reconstruct the touching trajectories while precisely analyzing the hovering part in order to position the next touching trace correctly. This relies on a Mixture-Of-Experts (MOE) approach. The first expert is dedicated for the pencil touch, and is named touching expert model. The second one is dedicated for the hovering pen trajectory, and is named hovering expert model. We improve on the learning of each of these experts based on additional context or specific examples. In addition we introduce a novel public benchmark dataset, to enable future research and comparisons in the field of handwriting reconstruction. The results demonstrates a significant enhancement compared to its primary competitors.
