Guile Wu

AI researcher; PhD in deep learning and computer vision.

I am an AI researcher with a strong passion for conducting cutting-edge AI research and applying them to real-world problems. I received my PhD degree from Queen Mary University of London, London, UK, under the supervision of Prof. Shaogang (Sean) Gong. I have broad research experience spanning deep learning, 3D reconstruction, generative AI, and autonomous driving. I have published some papers in top-tier AI venues and several patents. See DBLP or Google Scholar for more.


E-mail: guile [DOT] wu [AT] outlook [DOT] com

Area of Expertise:

My recent research focuses on exploring generative AI (particularly diffusion models, vision/language foundation models) for autonomous driving scene generation (e.g., world models), as well as neural rendering and 3D reconstruction (e.g., 3D Gaussian splatting, NeRF, SDF) for autonomous driving scene reconstruction and object simulation.

In addition, I have broad knowledge in deep learning, e.g., transfer learning, knowledge distillation, incremental learning, federated learning, unsupervised learning, domain adaptation, domain generalization, object detection, object re-identification, etc.

Selected Publication/Preprint (DBLP or Google Scholar):

  • "UniGaussian: Driving scene reconstruction from multiple camera models via unified gaussian representations",
    Y. Ren, G. Wu, R. Li, Z. Yang, Y. Liu, X. Chen, T. Cao, B. Liu,
    arXiv preprint, 2024.
  • "Vqa-diff: Exploiting vqa and diffusion for zero-shot image-to-3d vehicle asset generation in autonomous driving",
    Y. Liu, Z. Yang, G. Wu, Y. Ren, K. Lin, B. Liu, Y. Liu, J. Shan,
    ECCV, 2024.
  • "Towards universal LiDAR-based 3D object detection by multi-domain knowledge transfer",
    G. Wu, T. Cao, B. Liu, X. Chen, Y. Ren,
    ICCV, 2023.
  • "Mv-deepsdf: Implicit modeling with multi-sweep point clouds for 3d vehicle reconstruction in autonomous driving",
    Y. Liu, K. Zhu, G. Wu, B. Liu, Y. Liu, J. Shan,
    ICCV, 2023.
  • "Striking a balance between stability and plasticity for class-incremental learning",
    G. Wu, S. Gong, P. Li,
    ICCV, 2021.
  • "Collaborative optimization and aggregation for decentralized domain generalization and adaptation",
    G. Wu, S. Gong,
    ICCV, 2021.
  • "Peer collaborative learning for online knowledge distillation",
    G. Wu, S. Gong,
    AAAI, 2021.
  • "Decentralised learning from independent multi-domain labels for person re-identification",
    G. Wu, S. Gong,
    AAAI, 2021.
  • "Generalising without forgetting for lifelong person re-identification",
    G. Wu, S. Gong,
    AAAI, 2021.
  • "Tracklet self-supervised learning for unsupervised person re-identification",
    G. Wu, X. Zhu, S. Gong,
    AAAI, 2020.