1 paper
Zikang Liu, Longteng Guo, Handong Li +7
Real-time understanding of continuous video streams is essential for interactive assistants and multimodal agents operating in dynamic environments. However, most existing video re…