Research on Simulation and Visualization of Paint Materials in Gongbi Painting Based on Deep Learning
Main Article Content
Abstract
Gongbi painting, as an important category of traditional Chinese painting, relies on pigment materials (such as the heavy texture of mineral pigments and the transparency of plant pigments) and color setting techniques (such as “three alum and nine dye” and “multi-layer overlay”) as its artistic charm. However, traditional creation is time-consuming and laborious, and digital simulation techniques are difficult to accurately reproduce its physical characteristics and visual effects. The existing simulation research on Gongbi painting pigments relies heavily on traditional physical modeling, which has problems such as low accuracy in material restoration, stiff color overlay effects, significant differences between visual presentation and actual painting, and difficulty in balancing simulation efficiency and artistic authenticity. To this end, this article first sorts out the classification and characteristics of Gongbi painting pigments to clarify the core simulation requirements, and then constructs a pigment material simulation model based on deep learning. Combining Kubelka Munk color optics theory and Conditional Generative Adversarial Network (CGAN), three core modules are designed: pigment characteristic extraction, multi-layer overlay simulation, and visual rendering. Finally, the performance of the model is verified through comparative experiments. The experiment selected 10 typical gongbi painting pigments (5 mineral pigments, 5 plant pigments) and compared our model with traditional physical modeling methods and existing deep learning methods. The results showed that our model improved the accuracy of pigment texture restoration and color overlay similarity by more than 30%, and the visualization rendering speed increased by 50%. It can accurately reproduce the artistic effects of gongbi painting such as “thin color multi coating” and “blending gradient”, and has good real-time interactivity. Measured refractive, scattering and absorption parameters also make the model relevant to optical material visualization.
Downloads
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
References
W. Jiang, Z. Wang, J. S. Jin, et al., “DCT-CNN-based classification method for the Gongbi and Xieyi techniques of Chinese ink-wash paintings,” Neurocomputing, vol. 330, no. FEB.22, pp. 280-286, 2019, doi: 10.1016/j.neucom.2018.11.003.
X. Shi, Y. Li, J. Li, et al., “Audio-Visual Aware Foveated Rendering,” Visualization and Computer Graphics, IEEE Transactions on, vol. 31, no. 10, pp. 7711-7726, 2025, doi: 10.1109/TVCG.2025.3554737.
J. V. Krishna, P. Singh, R. Nagaraju, et al., “Design of an Improved Method for Visual Rendering in the Metaverse Using CIEM and MSRANet,” Access, IEEE, vol. 13, pp. 22176-22196, 2025, doi: 10.1109/ACCESS.2025.3532634.
A. I. Lysykh, D. D. Zhdanov, and M. I. Sorokin, “Method of Geometry Reconstruction from a Set of RGB Images Using Differentiable Rendering and Visual Hull,” Programming and Computer Software, vol. 51, no. 3, pp. 150-159, 2025, doi: 10.1134/S0361768825700033.
A. Singh, “Real-Time Generative AI in Game Texture Rendering: A Systematic Analysis of Visual Enhancement Technologies,” International Journal of Scientific Research in Computer Science, Engineering and Information Technology, vol. 10, no. 6, pp. 775-784, 2024, doi: 10.32628/CSEIT241061112.
A. F. Yulia, Zulkifli, P. Bintoro, et al., “Comparison of EEVEE and Cycles Rendering Performance in Blender 3.5 in the Context of Interactive Visuals for 3D Animation,” Jurnal Penelitian Pendidikan IPA, vol. 10, no. 7, pp. 4086-4091, 2024, doi: 10.29303/jppipa.v10i7.5910.
Y. Peng, “Appearance design of art exhibits combined with computer vision rendering technology,” International Journal of Information and Communication Technology, vol. 26, no. 22, pp. 1-22, 2025, doi: 10.1504/IJICT.2025.146910.
A. Dinita, C. M. Rosca, M. Tanase, et al., “A Comprehensive Review on Bridging the Research Gap in AI-Driven Material Simulation for FRP Composites,” Computer Modeling in Engineering & Sciences, vol. 144, no. 7, pp. 147-199, 2025, doi: 10.32604/cmes.2025.066276.
G. Zhang, G. Guo, Y. Gong, et al., “Failure Mechanism of Giant-Thick Strata Characterized by Weak Cementation Based on Superimposed Similar Material Simulation,” KSCE Journal of Civil Engineering, vol. 26, no. 8, pp. 3355-3369, 2022, doi: 10.1007/s12205-022-1423-4.
Y. Li, S. Y. Kim, S. Shaw, et al., “Material Property Simulation for Advanced Packaging,” Journal of Electronic Materials, vol. 55, no. 2, pp. 1664-1673, 2026, doi: 10.1007/s11664-025-12636-6.