Handwriting Trajectory Recovery with Diffusion Models
For researchers and practitioners in handwriting analysis and document forensics, this work provides a novel approach to stroke recovery that outperforms existing methods and shows cross-script generalization.
The paper introduces the first diffusion-model-based framework for handwriting trajectory recovery from offline images, achieving substantial improvements in temporal similarity (DTW/LDTW) and shape fidelity (AIoU) over prior methods like PEN-Net and Cross-VAE on CASIA-OLHWDB.
Recovering online pen trajectories from offline handwriting images, often referred to as handwriting trajectory recovery (stroke recovery), is an offline-to-online conversion task with applications in stroke-level editing and forensic analysis. We propose, to the best of our knowledge, the first diffusion-model-based framework for this task. Our method formulates trajectory recovery as image-conditioned generation and uses a denoising diffusion model to sample pen trajectories consistent with the observed ink trace. Through extensive quantitative evaluations on CASIA-OLHWDB (1.0-1.1), we verify that the proposed approach enables accurate recovery even for complex multi-stroke characters, substantially improving both temporal similarity (DTW/LDTW) and shape fidelity (AIoU) over representative prior methods such as PEN-Net and Cross-VAE. We further show that the model captures general stroke-order tendencies and generalizes to classes unseen during training, exemplified by cross-script transfer: a model trained on Chinese characters can recover reasonable stroke orders for Latin letters to some extent.