Xi'an Chang'an two hundred ten National Road
Xian
China
Xidian University
Skeleton-based Action Recognition, Transformer, Position Encoding, Multi-Scale Representation
Multimodal discussion summarization, Proactive agents, Multimodal knowledge representation, Knowledge acquisition
Multi-modal representation, referring expression segmentation, Segment Anything
Video understanding, coarse-to-fine, spatiotemporal exploration, agentic system