Dongdong Chen

Dongdong Chen 陈东东

Machine Learning | Imaging | Inverse Problems

Assistant Professor
Heriot-Watt University, Edinburgh, UK
d.chen@hw.ac.uk

I am an Assistant Professor of Mathematical and Computer Sciences at Heriot-Watt University in Edinburgh. My research lies at the intersection of machine learning and image processing. My current focus is on inverse problems in computational imaging and vision, through the lens of self-supervised learning, generative modelling, and foundation models. I pioneered groundtruth-free machine imaging (equivariant imaging). I am a lead developer of deep inverse, a popular library for solving inverse imaging problems with deep learning.

I held a postdoctoral researcher position at the University of Edinburgh from 2018 to 2022. I then held an associate researcher position at Sichuan University until 2023, where I previously received my PhD in 2017.

Openings: We are looking for motivated PhD students (full scholarship available), please see here for details.

news

Aug 05, 2024	Invited talk at the IMSI Workshop on Computational Imaging, University of Chicago, USA.
May 02, 2024	Our paper on foundation models for image fusion (FILM) is accepted at ICML’24.
Feb 27, 2024	Our paper on equivariant image fusion (EMMA) is accepted at CVPR’24.
Jan 09, 2024	Invited talk at the IMS Young Mathematical Scientist Forum – Applied Math, National University of Singapore (NUS), Singapore.

selected publications

ICCV OralFeatured

Equivariant Imaging: Learning Beyond the Range Space

Dongdong Chen, Julián Tachella , and Mike E Davies

In IEEE/CVF International Conference on Computer Vision (ICCV) , 2021

Oral Presentation [Featured in ICCV Daily]

arXiv Bib HTML PDF Supp Video Code

@inproceedings{chen2021equivariant,
  title = {Equivariant Imaging: Learning Beyond the Range Space},
  author = {Chen, Dongdong and Tachella, Juli\'an and Davies, Mike E},
  booktitle = {IEEE/CVF International Conference on Computer Vision (ICCV)},
  award2 = {Featured},
  further_highlight_with_url = {<a href="https://www.rsipvision.com/ICCV2021-Thursday/14/"> [Featured in ICCV Daily]</a>},
  year = {2021}
}

JMLR

Sensing Theorems for Unsupervised Learning in Linear Inverse Problems

Julián Tachella , Dongdong Chen, and Mike Davies

Journal of Machine Learning Research (JMLR), 2023

arXiv Bib PDF

@article{tachella2022jmlr,
  title = {Sensing Theorems for Unsupervised Learning in Linear Inverse Problems},
  author = {Tachella, Juli\'an and Chen, Dongdong and Davies, Mike},
  journal = {Journal of Machine Learning Research (JMLR)},
  volume = {24},
  number = {39},
  pages = {1--45},
  year = {2023}
}

IEEE SPM

Imaging With Equivariant Deep Learning: From unrolled network design to fully unsupervised learning

Dongdong Chen, Mike E Davies , J Matthias Ehrhardt , Carola-Bibiane Schönlieb , Ferdia Sherry , and Julián Tachella

IEEE Signal Processing Magazine, 2023

arXiv Bib PDF

@article{chen2022spm,
  title = {Imaging With Equivariant Deep Learning: From unrolled network design to fully unsupervised learning},
  author = {Chen, Dongdong and Davies, Mike E and Ehrhardt, J Matthias and Sch\"onlieb, Carola-Bibiane and Sherry, Ferdia and Tachella, Juli\'an},
  journal = {IEEE Signal Processing Magazine},
  volume = {40},
  number = {1},
  pages = {134-147},
  year = {2023}
}

Awards and Honors

MICCAI-BIA'18 Best Paper Nomination, 2018
Sichuan University Excellent Graduate, 2017
SCF Best Student Paper Award, 2017
IES Best Student Paper Award, 2014
National Scholarship, 2014