AI-Driven Semantic Re-encoding and Collaborative Design for the Holographic Presentation of Cultural Heritage

This paper systematically studies the "collaborative semantic redesign of holographic images of cultural holography using AI." Addressing the challenges of high reflectivity, complex decorative patterns, and structural incompleteness in the 3D reconstruction of porcelain and utensil artifacts, a holographic re-expression model integrating artificial intelligence (AI) is proposed. Accurate geometric modeling and color correction are achieved through multi-view imaging, hyperspectral scanning, and structured light data acquisition. NeRF reconstruction, rotational/axisymmetric constraints, and a Poisson fusion algorithm are combined to complete defect filling and topological restoration. A holographic algorithm based on deep learning and frequency domain optimization is further employed to generate printable computer-generated holograms (CGHs). This is validated using both holography color holography (Denisyuk) and digital direct write holography (DWDH) approaches, achieving an end-to-end closed loop from data to finished product. Experimental results demonstrate that AI-driven geometric and texture restoration, spectral mapping, and artifact suppression effectively reduce chromatic aberration (ΔE ≈ 1.8), enhance clarity, depth, and semantic usability, and shorten production cycles by approximately 30%. This approach provides a new technological paradigm for cultural relic display and digital humanities research.