Matt Dutson

Computer Sciences PhD Candidate at UW—Madison

dutson@wisc.edu

CV | GitHub | WISION Lab

Summary

I am a computer sciences PhD student at the University of Wisconsin—Madison, advised by Mohit Gupta. Broadly, my research interests include computational imaging, vision, graphics, and machine learning. Currently, I am working on efficient and stable video perception, and on passive single-photon imaging.

I will be graduating this year and am seeking a full-time position. I am primarily interested in solving real, meaningful problems, as opposed to pursuing technical novelty for its own sake. I am open to less traditional career paths and significant departures from my current research—if the vision is sound.

See my CV for a complete education history, work history, and list of publications. I have worked with many languages and technologies, although I have the most experience with the scientific Python stack (Linux, Bash, Git, Python, NumPy, and PyTorch). More important than any particular skill, I am eager to learn and not afraid of reading the documentation.

Selected Publications

• Adapting Frame-Based Networks for Stable and Robust Video Inference
  In Review
  Matthew Dutson, Nathan Labiosa, Yin Li, and Mohit Gupta
  
• Generalized Event Cameras
  2024 Conference on Computer Vision and Pattern Recognition (CVPR)
  Varun Sundar^*, Matthew Dutson^*, Andrei Ardelean, Claudio Bruschini, Edoardo Charbon, and Mohit Gupta
  Paper | Webpage
• Eventful Transformers: Leveraging Temporal Redundancy in Vision Transformers
  2023 International Conference on Computer Vision (ICCV)
  Matthew Dutson, Yin Li, and Mohit Gupta
  Paper | Webpage | PyTorch Code
• Spike-Based Anytime Perception
  2023 Winter Conference on Applications of Computer Vision (WACV)
  Matthew Dutson, Yin Li, and Mohit Gupta
  Paper | Webpage | TensorFlow Code
• Event Neural Networks
  2022 European Conference on Computer Vision (ECCV)
  Matthew Dutson, Yin Li, and Mohit Gupta
  Paper | Webpage | PyTorch Code | TensorFlow Code

Other Projects

Some of these projects are a few years old (e.g., I wrote the cosmic ray simulation code in 2017). For a recent code sample, see this repo containing the code for our Eventful Transformers paper.

• Simulating atmospheric cosmic ray showers
• Simulating images of collagen fibers
• Visualization tools for image classifiers
• YOLO object detection in TensorFlow
• Post-disaster building damage detection

Personal Contact

My social media accounts are all deleted or inactive. The best way to contact me is through email or text message (Signal, iMessage, or RCS). Email me if you need my phone number.

I sometimes send out a personal email newsletter. Message me if you'd like to be added to my mailing list.

About

I grew up in a small town in northern Utah and earned a bachelor's degree in physics from the University of Utah. I'm married to a good egg named Itzel who has given me a new perspective on what matters. We have a dog named Marzipan and a gecko named Monty. In my free time, I enjoy walking around outside, playing board games with my siblings, and building things (not necessarily software).