With a passion for bringing the latest know-hows to companies that can make a change
Full Resume
I've finished bachelors and masters degree from UBC and have worked with Mott Macdonald Canad Ltd. Working with engineers, and academics in computer vision, graphics and simulation, I developed a taste for works that leveraged your large scale software developement, but also incorporated math, physics and data sciences. I believe in thinking of the world in systems, and knowing when and how you can make them better. Thanks for coming here, you're welcome to look at some my projects. Here are the project topics you're likely to find in there:
Thesis and potential submission to SIGGRAPH Asia
Through modern differential simulators, it is now possible to backsolve deformations of objects for their material configurations. We leverage gaussians to track point deformations of non rigid material using sparse views and use them to FEM dynamics and volumetric material properities.
Submitted to 2025 IEEE Major Conference
(Conference details available upon request)
Materials used in real clothing exhibit remarkable complexity
and spatial variation due to common processes such as stitching,
hemming, dyeing, printing, padding, and bonding. Simulating
these materials, for instance using finite element methods, is
often computationally demanding and slow. Here we propose a
general framework for learning a simple yet efficient mass
mass-spring surrogate model that captures the effects of these
complex materials using only motion observations. Our method
achieves significantly faster training times, higher
reconstruction accuracy, and improved generalization to
scenarios where the cloth encounters novel dynamics.
2024 IEEE 18th International Conference on Automatic Face and Gesture Recognition
It is now possible to estimate 3D human pose from monocular images with off-the-shelf 3D pose estimators. However, many practical applications require fine-grained absolute pose information for which multi-view cues and camera calibration are necessary. Such multi-view recordings are laborious because they require manual calibration, and are expensive when using dedicated hardware. Our goal is full automation, which includes temporal synchronization, as well as intrinsic and extrinsic camera calibration
2019 BC AI Student Showcase Poster
My team's project of improving efficiency of pre-existing paper was selected for the BC AI shocase. Using Radial Basis Functions and Multi-class Logistic Regression. We were able to gat accuracy of about 96.4% in detecting states of motion when compared to the original paper.
Forked from fantastic work by Zeshun Zong et al., I built on top of it, a collision detection feature with non planar geomentries using line search and SDF colliders. I also added convinient rendering and initialization routines for the work. This allows for this method to be used on more complex environments.
Another Emerging Media Lab porject where we tried to use audio detection and FFT to map the frequencies to certain ranges and map them to colors. The colors where then used as visualize around the object emitting the sound to simulate the experience of synesthesia
Done with the team of Perigon Games, We designed a fun game where you had to identify the cat you were looking for by having conversations with other cats in the day care. It was a rewarding experience mostly for the team I got to work with. Highly recommend checking the out!