As an important carrier of Chinese culture, Calligraphy Seal Engraving carries deep historical and cultural connotations. In this study, we propose a generative model-based method for reconstructing calligraphic seal cutting strokes and designing 3D virtual display. The consistent point set drift (CPD) algorithm is used to match the stroke contour, combined with an autoregressive model to generate the calligraphic seal cutting texture, construct an asymmetric rhombic stroke model to characterize the strokes, and develop a 360-degree immersive 3D virtual display system. The experimental results show that the proposed method achieves 0.9437, 0.9531 and 0.9542 in SSIM, FM and S-measure, respectively, which are significantly better than the traditional OTSU method of 0.8165, 0.8596 and 0.8648. In the single dataset test, the proposed method obtains the best performance on 10 different inscriptions, among which the SSIM value of 0.9737 is achieved on the record of Miaoyanji Temple, and the SSIM value of 0.9542 is achieved on the record of Miaoyanji Temple. The SSIM value reaches 0.9765 and 0.9042 on the Shenzejun stele. The audience experience evaluation of the virtual display system shows an overall satisfaction score of 89.039, with the highest score of 89.912 for the narrative experience. Through the organic combination of stroke reconstruction and 3D virtual display, this method effectively realizes the digital protection and cultural inheritance of calligraphy seal cutting works, and provides a new technical path for the modernized display of traditional culture.
