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.
 

May 29th, 2026 – 12:00-1:00 pm, EC4-2101A

The recent advancements in artificial intelligence (AI) have been changing how we live, work, and dream, revolutionizing many industries through its versatile capabilities. Despite its many benefits, numerous challenges remain when integrating AI agents into our current workflows. No matter how advanced AI becomes, humans will remain in the loop at all times—focused more on goals, operating at higher levels of abstraction, and analyzing domain-specific circumstances. A major barrier to effective collaboration lies in the communication between humans and AI agents. In this talk, I will introduce my research, which takes a step toward addressing this gap using advanced interaction and visualization techniques to create harmony between humans and AI. I will share my perspective on the factors that influence human-AI communication, as well as showcase a range of example projects situated in various applications such as programming, information-seeking, and design.
 

May 22th, 2026 – 12:00-1:00 pm, EC4-2101A

My seminar focuses on 3D reconstruction of thin and strand-like structures such as hair, fur, fibres, and occluders. I first discuss direction-aware kernel support for neural implicit reconstruction of fuzzy geometry, then present a controlled 3D Gaussian Splatting study where thin occluders partially hide a target object. The goal is to understand how directional structure affects both representation quality and reconstruction capacity under occlusion.

Dr. Freda Shi

Personal website: https://mkoshkina.github.io/

Linkedin: https://www.linkedin.com/in/kseniiabuzko/

March 27th, 2026 – 12:00-1:00 pm, EC4-2101A

Traditionally, most studies of microbes involve bulk population measurements of, e.g. well-mixed populations grown in culture tubes. However, spatial distribution is known to play an important role in microbial ecology. Our group has developed a pipeline to characterize ecological interactions among microbial populations from time-lapse microscopy at single cell resolution. We collect images of mixed populations of ~1000 cells constrained to two dimensions, at a frequency of 3-5 minutes. Image processing (segmentation and object tracking) is implemented with CellProfiler and our custom software package TrackRefiner. We then use processed image data to calibrate agent-based models of cellular activity. These models allow us to assess hypotheses regarding cellular behaviour and build toward predictive model-based design of interventions in natural and engineered microbial systems.