Summary - F.748.72 (12/2025) - Requirements and framework of multimodal generative artificial intelligence (AI)-enabled multi-view transformation

Multimodal generative artificial intelligence (AI) can help to deal with different modalities (e.g., images, videos, laser points, map) and solve the problems of traditional AI-based multi-view transformation (MVT) systems, such as lack of capability to produce obstructed or blocked views and information, and deficient in prior information which is necessary to comprehend spatial relations. Recommendation ITU-T F.748.72 specifies the framework, use cases and requirements of a multimodal generative AI-enabled multi-view transformation (MEMVT) system, including information compensation, multi-view fusion and completion, multimodal temporal fusion and application functions. This Recommendation is used to augment the generation capability of multi-view transformation systems, thereby enhancing its precision and robustness, and improving the transformation quality and application effectiveness.