Qiushan Guo

slack.png

Graduate House, P501

The University of Hong Kong

I am a PhD student at The University of Hong Kong, supervisde by Prof. Yizhou Yu (ACM & IEEE Fellow) and Prof. Ping Luo (Associate Director of HKU Musketers Foundation Institute of Data Science). I am part of the HKU MMLab led by Prof. Ping Luo. During my PhD study, I was lucky to collaborate with several researchers in industry e.g. Dr. Sifei Liu and Dr. Hongxu Yin @ NVIDIA and Dr. Junjie Yan @ Sensetime.

My work has focused on probabilistic modeling of high-dimensional data, large vision-language model and the application of this technique to various domains. Specifically, I investigate the efficient neural network, such as dynamic routing and knowledge distillation. I also had experience in the semi-supervised and self-supervised learning of object detection.

One of my recent notable contributions involves RAPHAEL, a large-scale text-to-image generator featured in SenseMirage.

My research interest is solving challenging problems in computer vision and machine learning, including:

  • Diffusion model
  • Large vision-language model
  • Object detection
  • Knowledge distillation

If you are interested in my research or collaboration with me, please feel free to drop an email.

selected publications

  1. rgpt.gif
    Regiongpt: Towards region understanding vision language model
    Qiushan Guo , Shalini De Mello , Hongxu Yin , and 5 more authors
    CVPR, 2024
  2. EGC.png
    EGC: Image Generation and Classification via a Diffusion Energy-Based Model
    Qiushan Guo , Chuofan Ma , Yi Jiang , and 3 more authors
    ICCV, 2023
  3. KDCL.png
    Online knowledge distillation via collaborative learning
    Qiushan Guo , Xinjiang Wang , Yichao Wu , and 4 more authors
    CVPR Oral, 2020