AI RESEARCH

FlowNar: Scalable Streaming Narration for Long-Form Videos

arXiv CS.CV

ArXi:2606.00620v1 Announce Type: new Recent Large Multimodal Models (LMMs), primarily designed for offline settings, are ill-suited for the dynamic requirements of streaming video. While recent online adaptations improve real-time processing, they still face critical scalability challenges, with resource demands typically growing at least linearly with video duration. To overcome this bottleneck, we propose FlowNar, a novel framework for scalable streaming video narration.