October 2ed, 2026 – 12:00-1:00 pm, EC4-2101A

Modeling is a crucial component of research in glaciology for examining drivers of ice flow, ice mass loss into the ocean, and related sea level rise. The difficulty of accessing glaciers, particularly the bed, which lies under hundreds, if not thousands of meters of ice, means that models must fill a significant gap in our knowledge of glacial conditions. However, this also means that there are limited data available for model validation and, for model inputs, the datasets often contain large errors or omissions. In this seminar, I will introduce the current state of glacial modeling, including recent advances in coupling subglacial hydrology and ice dynamics. I will also discuss where the research gaps are, with focus on the data that will be required to fill those gaps.  
 

Linkedin: https://www.linkedin.com/in/yuhsi-li/

Linkedin: https://www.linkedin.com/in/zirui-iris-chen/

September 4th, 2026 – 1:00-2:00 pm, EC4-2101A

Recent advancements in sensing and deep learning have unlocked exciting possibilities for the visual analysis of human and animal motions in the physical 3D space. These innovations hold great potential for applications across diverse domains, including for example natural user interfaces, AR/VR, robotics, and gaming. In this talk, I will present the latest research progress in this rapidly evolving field including especially 3D human motion generation, pose tracking and shape reconstruction, and related tasks – highlighting key developments from the past few years as well as contributions from our own work.
 

June 19th, 2026 – 12:00-1:00 pm, EC4-2101A

This presentation describes the current state of the interdisciplinary Living Lab being developed on campus. The goal is to connect researchers across disciplines and make it easier to bring real environmental data into classes, undergraduate design projects, and research.

We will give a quick overview of the existing sensor setups in E2, RCH, and CPH, which have been collecting data for about 1 to 2 years. We will also discuss the planned expansion of the network, including new sensors to measure salt concentration in Laurel Creek, which is scheduled for deployment this summer.

Finally, we will introduce our new platform, DataHub.uwaterloo.ca, which brings together all data collected from campus sensors and citizen science into a single easy-to-use resource for both research and teaching. 
 

June 12th, 2026 – 12:00-1:00 pm, EC4-2101A

3D Gaussian Splatting (3DGS) has emerged as a leading method for photorealistic scene reconstruction from multi-view images, yet existing approaches treat reconstruction as a purely visual optimization problem, ignoring the physical laws that govern real-world scenes. This talk explores a bidirectional relationship between physics and vision within the 3DGS framework. In the forward direction, physical and geometric priors are introduced at each stage of the pipeline: upsampled point cloud initialization improves reconstruction quality without architectural changes, mesh-coupled Gaussian representations enable physics simulation of dynamic scenes, and differentiable rigid-body simulation provides trajectory supervision during object occlusion in 4D reconstruction. In the inverse direction, we examine how observed fracture behaviour in food materials can be used to recover latent material parameters via surrogate modelling and reinforcement learning, enabling novel simulation of physically plausible fracture dynamics. Together, these contributions demonstrate that neither physics nor vision alone is sufficient for faithful dynamic reconstruction and simulation, and that integrating the two disciplines yields measurable and qualitatively meaningful improvements across all stages of the pipeline.
 

June 5th, 2026 – 12:00-1:00 pm, EC4-2101A

Previous eye tracking research in sports has relied on hand annotated areas of interest to label data. This bottlenecked the number of participants and trials that could be used for research leading to small sample sizes and expensive experimental set ups. Using RF-DETR and controlled environments, it is possible to automate ball detection with reasonable accuracy speeding up the analysis and increasing the sample size. The presentation will show how I automated data collection and how the automated ball detection helps with analyzing data in baseball and volleyball.