Yiwen Gu

Yiwen Gu is a third year PhD student at Boston University in the Image and Video Computing Group. She is advised by Professor Margrit Betke. Her interests lie in the intersection of computer vision and human computer interfaces.

yiweng [at] bu.edu | CV | Google Scholar

Teaching Experience

Yiwen has been a teaching fellow for

  • CS 640: Artificial Intelligence, Fall 2020
  • Research in Science & Engineering Program (RISE), Summer 2020
  • CS 111: Introduction to Computer Science, Spring 2020
Working Experience
  • Software Engineer Intern, SAIL, Hariri Institute for Computing, BU
  • Backend Developer Intern, MIDA. Inc
  • the NSF Doctoral Consortium (DC) award, PETRA
  • Invited participant for the Grad Cohort Workshop of the CRA-W
Research Projects

Animal Pose 3D Reconstruction

Spatial cognition is a critical component of intelligent behavior. How does a trained animal recall and navigate between known goals is a question that attracts not only neuroscientists but also scientists of robotics. In this collaborative study, we seek to map rat's brain activity and its behavior. Our goal in the current stage is to develop a reliable model that can track various body parts of a rat in a 3D coordinate system.


ExerciseCheck is a scalable, accessible platform designed and developed for the remote monitoring and evaluation of physical therapy. It is a modular system that incorporates machine learning techniques with contemporary web technologies to enable a user-friendly experience for patients and physical therapists.
The project has gone through 3 phases. Phase I is the initial development. We tested the system in our CS lab with 2 healthy subjects and 2 physical therapists. In the Phase II we tested the system with 5 Parkinson patients in a rehabilitation clinic in the Sargent College. We evaluated the effectiveness of the visual feedback on the performance and the quantitative analysis and received users' feedback. In the Phase III we invited 3 Patients and deployed the system in their homes for long-term user feedback.

Predicting Aphasia Recovery

The potential recovery of post-stroke aphasia is highly variable and the rehabilitation outcomes are difficult to predict. This interdisciplinary collaboration builds on data collected as part of a large set of behavioral and brain variables in patients with post-stroke aphasia, charting the course of recovery associated with therapy across language domains and examining the basis of neuroplasticity.
In a pilot study, we created and tested a predictive framework based on a subset of the data collected and developed machine-learning algorithms that take as input a complex set of brain and behavioral features to classify and predict the participants' responsiveness to therapy.

Professional Services

    Member of Program Committee:


    Reviewer for Journals/Conferences:

  • International Journal of Computer Vision
  • ECCV/ACVR 2020-2022, the 8th-10th international workshop on Assistive Computer Vision and Robotics in conjunction with the European Conference on Computer Vision.
  • PETRA 2020-2021, ACM International Conference on PErvasive Technologies Related to Assistive Environments
  • Journal of Pattern Analysis and Applications (PAAA)
  • Journal of PLOS ONE