Guangming Shi

Ditch slow hyperspectral image processing: FUN achieves real-time object detection by jointly learning reconstruction and detection in a single, efficient network.

Direct generative decoding from semantics to images reduces hallucinations and accelerates inference speed by over eight times.

VLMs struggle to connect the dots between dynamic drone footage and satellite imagery, highlighting a critical gap in their spatial reasoning abilities.

By distilling visual foundation models, this work achieves a significant leap in event stream representation learning, surpassing prior methods in generalization, data efficiency, and transferability.

Semantic communication doesn't have to be computationally expensive: this paper introduces a framework that explicitly trades off rate, semantic distortion, and model complexity, enabling efficient system design.