Kerui Chen

Beyond the Single Camera: Agentic Multi-View Reasoning in Sports Video Understanding

Kerui Chen, Jinglu Wang, Xiaoyi Zhang, Yan Lu

Recent Multimodal Large Language Models (MLLMs) achieve strong performance on single-view video understanding benchmarks. However, sports videos involve dense occlusion, rapid motion, and complex interactions that are difficult to resolve from a single viewpoint. In practice, spo…Read more