Exploration of the cultural attributes of Chinese character sculpture using machine learning technology
Abstract
The article employs machine learning, specifically the CLIP (Contrastive Language-Image Pretraining) model, to analyze Chinese character sculptures’ cultural attributes. It overcomes challenges in multi-dimensional data processing and high digitization costs. The process involves normalizing sculpture images, using FastText for vector representations of Chinese characters, and mapping text to the same embedding space as images for word embedding. The CLIP model, through unsupervised training, minimizes the negative logarithmic likelihood loss between image and text embeddings to establish cultural attribute representations. Key findings include the CLIP model’s improved performance over the M3 model, with a 5.4% higher average AUC. The model demonstrates high efficiency and accuracy, evident in its low RMSE (0.034) and MAE (0.025) and fast analysis time of 182 ms. This approach effectively and accurately analyzes the cultural attributes of Chinese character sculptures, addressing existing research gaps.
Keywords
Full Text:
PDFReferences
1. Koutsabasis P, Vosinakis S. Kinesthetic interactions in museums: conveying cultural heritage by making use of ancient tools and (re-) constructing artworks. Virtual Reality. 2017, 22(2): 103-118. doi: 10.1007/s10055-017-0325-0
2. Guo X. Analysis on the Style and Evolution of the Sculpture Art in Shanxi Merchants Courtyard. Highlights in Art and Design. 2023, 3(2): 57-59. doi: 10.54097/hiaad.v3i2.10229
3. Liang W, Tadesse GA, Ho D, et al. Advances, challenges and opportunities in creating data for trustworthy AI. Nature Machine Intelligence. 2022, 4(8): 669-677. doi: 10.1038/s42256-022-00516-1
4. Li H, Wang W, Li Q, et al. A novel minimum-time feedrate schedule method for five-axis sculpture surface machining with kinematic and geometric constraints. Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture. 2018, 233(5): 1483-1499. doi: 10.1177/0954405418780167
5. Barreau JB, Jouneau J, Charlet C, et al. Digitization, Virtual Reality and Robotic Sculpture for the Preservation and Enhancement of the Public Heritage of the Sculpted Rocks of Rothéneuf. Journal on Computing and Cultural Heritage. 2022, 15(4): 1-21. doi: 10.1145/3522595
6. García-Molina DF, López-Lago S, Hidalgo-Fernandez RE, et al. Digitalization and 3D Documentation Techniques Applied to Two Pieces of Visigothic Sculptural Heritage in Merida Through Structured Light Scanning. Journal on Computing and Cultural Heritage. 2021, 14(4): 1-19. doi: 10.1145/3427381
7. Bovcon N. Virtual museums: interpreting and recreating digital cultural content. Neohelicon. 2021, 48(1): 23-38. doi: 10.1007/s11059-021-00582-1
8. Rani A. Digital Technology: It’s Role in Art Creativity. Journal of Commerce & Trade. 2018, 13(2): 61. doi: 10.26703/jct.v13i2-9
9. Christidou D, Pierroux P. Art, touch and meaning making: an analysis of multisensory interpretation in the museum. Museum Management and Curatorship. 2018, 34(1): 96-115. doi: 10.1080/09647775.2018.1516561
10. Zhilin M, Savchenko S, Hansen S, et al. Early art in the Urals: new research on the wooden sculpture from Shigir. Antiquity. 2018, 92(362): 334-350. doi: 10.15184/aqy.2018.48
11. Cialone C, Tenbrink T, Spiers HJ. Sculptors, Architects, and Painters Conceive of Depicted Spaces Differently. Cognitive Science. 2017, 42(2): 524-553. doi: 10.1111/cogs.12510
12. Yang J. Analysis of the Clever Application of Sculpture Language in Modern Sculpture. Tiangong. 2018, (1): 16-17.
13. Scott DA. Ancient Marbles: Philosophical Reflections on the Restoration of Sculpture. Studies in Conservation. 2022, 68(4): 388-406. doi: 10.1080/00393630.2022.2049032
14. Fraser H. Grief encounter: the language of mourning in fin-de-siècle sculpture. Word & Image. 2018, 34(1): 40-54. doi: 10.1080/02666286.2017.1333880
15. Barata M. The Sculpture of Black Origin in Brazil. Art in Translation. 2022, 14(1): 96-102. doi: 10.1080/17561310.2022.2046532
16. Mihai Ionut R. Technology and imagination in contemporary art. aspects of modern sculptural object. Limba Si Literatura–Repere Identitare in Context European. 2018, 22(22): 276-284.
17. Heginbotham A, Erdmann R, Hayek LAC. The dating of French gilt bronzes with ED-XRF analysis and machine learning. Journal of the American Institute for Conservation. 2018, 57(4): 149-168. doi: 10.1080/01971360.2018.1515389
18. Jabbarov RR. Patterns in applied art of the Uzbek folk. European Journal of Arts. 2023, (1): 11-14. doi: 10.29013/eja-23-1-11-14
19. Pulham P. The Sculptural Body in Victorian Literature. Edinburgh University Press, 2020. doi: 10.1515/9780748693436
20. Ullah I, Soomro MA, Mudassar Zulfiqar. A Review of Archaeological Reports and Literature on the Gandhara Sculpture Collection of the Royal Ontario Museum. Academic Journal of Social Sciences (AJSS). 2020, 4(3): 377-403. doi: 10.54692/ajss.2020.04031212
21. Zhao J, Liu X, Luo W, et al. Research on multimodal search tools for military image resources based on CLIP model. Chinese Journal of Medical Library and Information. 2022, 31(8): 14-20.
22. Baldrati A, Bertini M, Uricchio T, et al. Composed Image Retrieval using Contrastive Learning and Task-oriented CLIP-based Features. ACM Transactions on Multimedia Computing, Communications, and Applications. 2023, 20(3): 1-24. doi: 10.1145/3617597
23. Chen Z, Du H, Wu Y, et al. Cross modal video clip retrieval based on visual text relationship alignment. Chinese Science: Information Science. 2020, 50(6): 862-876. doi: 10.1360/SSI-2019-0292
24. Du P, Li X, Gao Y. A Review of Research on Multimodal Visual Language Representation Learning. Journal of Software Science. 2020, 32(2): 327-348.
25. Zhang L, Zhang L, Yuan Q. Remote sensing big models: progress and prospects. Journal of Wuhan University (Information Science Edition). 2023, 48(10): 1574-1581.
26. Wang C. Artificial Intelligence Driven Digital Image Art Creation: Methods and Case Analysis. Journal of Intelligent Science and Technology. 2023, 5(3): 406-414.
27. Chefer H, Alaluf Y, Vinker Y, et al. Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Models. ACM Transactions on Graphics. 2023, 42(4): 1-10. doi: 10.1145/3592116
28. Jiang D, Ye M. Transformer network for cross modal text to image pedestrian recognition [J]. Chinese Journal of Image and Graphics. 2023, 28(5): 1384-1395.
29. Liu T, Wu Z, Chen J, Jiang Y. A multimodal pre training method for visual language understanding and generation. Journal of Software. 2022, 34(5): 1-11. doi: 10.13328/j.cnki.jos.006770
30. Gal R, Patashnik O, Maron H, et al. StyleGAN-NADA. ACM Transactions on Graphics. 2022, 41(4): 1-13. doi: 10.1145/3528223.3530164
31. Huang Z, Bianchi F, Yuksekgonul M, et al. A visual–language foundation model for pathology image analysis using medical Twitter. Nature Medicine. 2023, 29(9): 2307-2316. doi: 10.1038/s41591-023-02504-3
32. Avrahami O, Fried O, Lischinski D. Blended Latent Diffusion. ACM Transactions on Graphics. 2023, 42(4): 1-11. doi: 10.1145/3592450
33. Zhou K, Yang J, Loy CC, et al. Learning to Prompt for Vision-Language Models. International Journal of Computer Vision. 2022, 130(9): 2337-2348. doi: 10.1007/s11263-022-01653-1
34. Gal R, Arar M, Atzmon Y, et al. Encoder-based Domain Tuning for Fast Personalization of Text-to-Image Models. ACM Transactions on Graphics. 2023, 42(4): 1-13. doi: 10.1145/3592133
35. Zheng Z, Zheng L, Garrett M, et al. Dual-path Convolutional Image-Text Embeddings with Instance Loss. ACM Transactions on Multimedia Computing, Communications, and Applications. 2020, 16(2): 1-23. doi: 10.1145/3383184
DOI: https://doi.org/10.32629/jai.v7i4.1471
Refbacks
- There are currently no refbacks.
Copyright (c) 2024 Zhen Luo
License URL: https://creativecommons.org/licenses/by-nc/4.0/