COLLABORATIVE AI SYSTEMS SUPPORTING DESIGN TEAMS IN PRODUCING LARGE-SCALE VISUAL ART PROJECTS

Such projects as large-scale visual arts, such as installations in the city, murals, and online exhibitions, require coordination among multidisciplinary design teams as never before. Conventional collaborative operations experience a major challenge in the efforts of ensuring consistency of arts, handling real-time contributions by the distributed workforce, and scaling artistic operations. The presented paper presents a new collaborative AI model which is specifically aimed to assist design teams to create the large-scale visual artworks. The system suggested above combines a multi-layer architecture that includes the data acquisition layer, AI processing layer, collaboration management layer, and immersive visualization layers. We use the most up-to-date generative models such as diffusion transformers and multi-agent artificial intelligences to facilitate the creation of co-creativity between human artists and intelligent agents. The system has built-in high-level synchronization techniques, version control systems, and customized recommendation systems based on creative processes. Experimental analysis indicates that team productivity (42%) and artistic coherence (87% quality score) and workflow efficiency (35 reduction in iteration time) were highly improved over traditional and semi-automated ones. Precision, recall, and F1-score in content generation tasks are found to be 0.91, 0.89, and 0.90 respectively in the quantitative analysis. Scalability tests assure real-time performance of a team with up to 50 concurrent users with under 200ms latency. This study building blocks provides the ground work of human-AI joint creativity in scale-artistic production.