Canhui Xu

I am a faculty member at Qingdao University of Science and Technology and a visiting scholar at Paderborn University, working on multimodal large model inference acceleration and embodied intelligence applications, with particular interest in document image understanding, inference acceleration, robotic reinforcement learning, and resource-constrained model deployment.

Previously, I completed my joint Ph.D. training at Imperial College London and my postdoctoral research at Peking University, and was a visiting scholar at Arizona State University. I have published nearly 30 academic papers in well-known international journals such as IEEE Transactions on Multimedia and IEEE Transactions on Image Processing, as well as mainstream international conferences, and have led research projects including those funded by the National Natural Science Foundation of China and commissioned enterprise R&D projects. I have also obtained 4 authorized invention patents and developed fully independently owned intellectual property systems such as the “Document Structured Information Extraction System” and the “Medical Image Segmentation System.”