Anisotropic Gaussian Pressure-Based Algorithm for Aerial Handwriting Trajectory Modeling

This research proposes a dynamic handwriting trajectory generation method based on anisotropic Gaussian pressure, integrating velocity, pressure information, and spatial diffusion characteristics of finger motion trajectories to establish a high-fidelity handwriting simulation model. To address the limitations of conventional methods-including high error rates, low fidelity, and insufficient dynamic adaptability-we innovatively design a multimodal pressure-sensitive algorithm. By dynamically adjusting the aspect ratio of the Gaussian kernel in real time according to the trajectory tangent direction and velocity, the algorithm simulates the anisotropic diffusion properties of ink, achieving a Bhattacharyya Distance Variance Score of 20.05, which significantly outperforms traditional approaches. Validation experiments confirm that the generated trajectories exhibit superior comprehensive performance compared to existing methods. This work provides a high-fidelity solution for handwriting interaction simulation in intelligent education and virtual reality handwriting systems.