Abstract: This paper presents a novel approach for interactively recovering dynamic writing processes from static Chinese calligraphy images.After collecting some ink samples from real calligraphic works, we first propose a statistic-based method to model the footprints between brush bundles and the paper. Using pen input devices,we then propose an efficient method for generating stroke trajectories from Chinese character images.Finally,we reconstruct a writing animation by calculating footprint series along given stroke trajectories.The proposed method has many potential applications,such as Chinese character teaching and digital media applications.