Zheyuan (David) Liu

Zheyuan (David) Liu

PhD Student, Deep Learning

Australian National University

Acton, Canberra



I am a PhD student at the Australian National University, supervised by Prof. Stephen Gould and working with Dr. Damien Teney (Idiap Research Institute | AIML). My research primarily focuses on vision-and-language reasoning in deep learning, with recent work on the task of composed image retrieval and generation.

To contact me: zheyuan.liu [at] anu.edu.au



Interests
  • Vision-and-Language
  • Visual Reasoning
  • Language-conditioned Image Retrieval
  • Image Generation
Education

  • PhD in Deep Learning, Computer Science, Current

    Australian National University

  • Bachelor of Engineering Hons (R&D), Electronics and Communication Systems, 2018

    Australian National University

Publications

Recent research

Zheyuan (David) Liu, Cristian Rodriguez-Opazo, Damien Teney, Stephen Gould
October, 2021 In ICCV 2021
Image retrieval on real-life images with pre-trained vision-and-language models
Image retrieval on real-life images with pre-trained vision-and-language models

We extend the task of composed image retrieval by introducing the first open-domain dataset along with a transformer based method.
PDF Cite Dataset Project Preprint (incl. supp. mat.)