Soumyabrata Dev (@soumyabratadev.bsky.social)

New paper in Multimedia Tools and Applications with @ucddublin.bsky.social: a multimodal hybrid fusion framework combining vision and text for billboard categorization in video, showing why multimodal AI outperforms single modality models. link.springer.com/article/10.1...

loading . . .

Multi-modal fusion for billboard categorization in video frames: A text and image-based approach - Multimedia Tools and Applications Traditional multimedia classification techniques rely on analyzing either its features or the associated annotated textual information. In this paper, we introduce a technique that leverages deep… https://link.springer.com/article/10.1007/s11042-026-21192-y