CV

Education, experience, and awards.

Contact Information

Name Heming Du (杜赫铭)
Professional Title Postdoctoral Research Fellow, The University of Queensland
Email heming.du@uq.edu.au
Website https://xiaobaishu0097.github.io

Professional Summary

Inclusive computer vision spanning pose estimation, adaptive 3D human mesh recovery, dense human parsing, embodied visual navigation, and vision-language models.

Experience

  • 2025 – Present

    Postdoctoral Research Fellow
    The University of Queensland
    INVITA Graindata Fusion project: multimodal agricultural data analytics, visualisation and statistical modelling to support crop management and yield prediction. Co-supervising PhD research on text-to-video retrieval and temporal moment retrieval.
  • 2024 – 2025

    Research Fellow (part-time)
    The University of Queensland
    Faculty of Engineering, Architecture and Information Technology (EAIT). Multimodal video understanding research.
  • 2022 – Present

    Research Scientist
    Follow Me AI Pty Ltd (FOME.ai)
    Computer vision for sport and Paralympic technology.
    • YouFor2032: athlete talent identification using pose estimation to support the Brisbane 2032 Olympics, with the Queensland Academy of Sport.
    • AQIRP (2024 to 2027): video analysis with AI to improve Paralympic classification, supported by keypoint frameworks, datasets, and pose estimation algorithms for athletes with limb differences.
  • 2020 – 2021

    PhD Researcher (industry collaboration)
    CSIRO (Commonwealth Scientific and Industrial Research Organisation)
    Coral reef monitoring with CSIRO Oceans and Atmosphere using weak supervision for 3D point cloud instance segmentation and underwater sensing over extended periods on the Great Barrier Reef.

Education

Awards

  • 2023
    First Prize: Image Classification of Nutrient Deficiencies in Winter Wheat and Winter Rye Challenge (ICCV 2023)
    University of Bonn

  • 2019
    First Prize (two tracks): 1st Mandarin Audio-Visual Speech Recognition Challenge, visual word-level recognition and visual keyword spotting
    Chinese Academy of Sciences

  • 2019
    2nd Runner-Up: CVPR 2019 AI City Challenge, City-Scale Multi-Camera Vehicle Re-Identification
    NVIDIA Corporation

  • 2019
    5th Place: CVPR 2019 AI City Challenge, City-Scale Multi-Camera Vehicle Tracking
    NVIDIA Corporation

  • 2019
    First Prize: 2nd Facial Micro-Expression Grand Challenge (FG 2019)
    Manchester Metropolitan University