WCCA-AK: A Multimodal Dataset of André Kim's Fashion Legacy for AI-Driven Cultural Heritage Research

초록

In this paper, we present WCCA-AK, a large-scale dataset of 3D scans and multi-view images capturing 100 haute couture garments by André Kim (1962-2010). The dataset supports tasks in 3D reconstruction, generation, and computational analysis, while serving as a bridge between computer vision research and cultural heritage preservation. To ensure responsible use, we establish an ethical framework that addresses the dual role of digital archives: faithfully documenting historical artifacts and enabling generative continuations of artistic vision. This duality reframes cultural preservation from static documentation into generative exploration, allowing computational methods to both honor historical legacy and inspire future creativity. WCCA-AK thus functions as (1) a high-fidelity archive of André Kim's work and (2) a foundation for studying how artistic style evolves and can be computationally extended while maintaining cultural authenticity. Looking ahead, we aim to expand the collection to 1,000 garments and extend the methodology to other designers. We invite researchers in computer vision, machine learning, and digital humanities to leverage WCCA-AK to advance both technical innovation and cultural understanding, ensuring that artistic legacies continue to inspire future generations. © 2025 IEEE.

키워드