Mikaela Angelina Uy
I am a fifth year PhD student at Stanford University advised by Leonidas Guibas. Broadly, my research interests are in 3D vision, geometry processing, graphics and machine learning. Specifically, I am interested in diving into different representations of 3D objects and scenes for various downstream tasks such as deformation, reconstruction, controllable generation and variation synthesis. I am particularly drawn to designing methods that connect classical techniques to learning-based approaches that are fundamentally-grounded and mathematically-inspired.
I am a returning research intern at Google in the Project Starline team, and was also previously a research intern in Adobe Research and Autodesk AI Lab. During my PhD, I am fortunate to have also closely collaborated with Ke Li and Minhyuk Sung, whose advice have also guided me to grow, appreciate and develop my taste in research.
Previously, I received my Bachelor's degree double majoring in Mathematics and Computer Science from the Hong Kong University of Science and Technology (HKUST) in 2017, and my Master's in Computing from the National University of Singapore in 2018. I then returned to HKUST as a Research Assistant for a year and had the pleasure of working with Sai-Kit Yeung, Binh-Son Hua and Duc Thanh Nguyen.
For highly-motivated students interested in projects related to NeRFs and/or geometry processing and shape analysis please fill out this google form. I am looking for potential collaborators to work on exciting and fundamental problems with!
Also check out a feature article for Women in Computer Vision from RSIP Vision here.
NeRF Revisited: Fixing Quadrature Instability in Volume Rendering
Mikaela Angelina Uy, George Kiyohiro Nakayama, Guandao Yang, Rahul Krishna Thomas, Leonidas Guibas, Ke Li
Advances in Neural Information Processing Systems (NeurIPS), 2023
DiffFacto: Controllable Part-Based 3D Point Cloud Generation with Cross Diffusion
George Kiyohiro Nakayama, Mikaela Angelina Uy, Jiahui Huang, Shi-Min Hu, Ke Li, Leonidas Guibas
IEEE International Conference on Computer Vision (ICCV), 2023
OptCtrlPoints: Optimizing Control Points for Biharmonic 3D Shape Deformation
Kunho Kim*, Mikaela Angelina Uy*, Despoina Paschalidou, Alec Jacobson, Leonidas Guibas, Minhyuk Sung
Pacific Graphics (Full Paper), 2023
SCADE: NeRFs from Space Carving with Ambiguity-Aware Depth Estimates
Mikaela Angelina Uy, Ricardo Martin-Brualla, Leonidas Guibas, Ke Li
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023
PartNeRF: Generating Part-Aware Editable 3D Shapes without 3D Supervision
Konstaninos Tertikas, Despoina Paschalidou, Boxiao Pan, Jeong Joon Park, Mikaela Angelina Uy, Ioannis Emiris, Yannis Avrithis, Leonidas Guibas
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023
Point2Cyl: Reverse Engineering 3D Objects from Point Clouds to Extrusion Cylinders
Mikaela Angelina Uy*, Yen-yu Chang*, Minhyuk Sung, Purvi Goel, Joseph Lambourne, Tolga Birdal, Leonidas Guibas
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022
Joint Learning of 3D Shape Retrieval and Deformation
Mikaela Angelina Uy, Vladimir G. Kim, Minhyuk Sung, Noam Aigerman, Siddhartha Chaudhuri, Leonidas Guibas
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021
Deformation-Aware 3D Model Embedding and Retrieval
Mikaela Angelina Uy, Jingwei Huang, Minhyuk Sung, Tolga Birdal, Leonidas Guibas
European Conference on Computer Vision (ECCV), 2020
LCD: Learned Cross-Domain Descriptors for 2D-3D Matching
Quang-Hieu Pham, Mikaela Angelina Uy, Binh-Son Hua, Duc Thanh Nguyen, Gemma Roig, Sai-Kit Yeung
AAAI Conference on Artificial Intelligence (AAAI), 2020 (Oral)
Revisiting Point Cloud Classification: A New Benchmark Dataset and Classification Model on Real-World Data
Mikaela Angelina Uy, Quang-Hieu Pham, Binh-Son Hua, Duc Thanh Nguyen, Sai-Kit Yeung
IEEE International Conference on Computer Vision (ICCV), 2019 (Oral)
- Apple Machine Learning Research (MLR), September 7, 2023, Towards Controllable 3D Content Creation by Leveraging Geometric Priors
- Google, July 12, 2023, NeRF Revisited: Fixing Quadrature Instability in Volume Rendering
- SFU Visual Computing and Robotics (VCR) Seminar, June 26, 2023, Towards Controllable 3D Content Creation by Leveraging Geometric Priors
- Structural and Compositional Learning on 3D Data, CVPR 2023 Workshop, June 18, 2023, Towards Controllable 3D Content Creation by Leveraging Geometric Priors
- KAIST, January 9, 2023, SCADE: NeRFs from Space Carving with Ambiguity-Aware Depth Estimates
- VinAI Seminar Series, July 22, 2022, Learning to Vary 3D Models for Universally Accessible 3D Content Creation
- Brown Vision Computing Seminar, April 11, 2022, Learning to Vary 3D Models for Universally Accessible 3D Content Creation
- Stanford G-Cafe, March 10, 2022, Point2Cyl: Reverse Engineering 3D Objects from Point Clouds to Extrusion Cylinders
- EECS Rising Stars, 2023
- Apple Scholars in AI/ML PhD Fellowship, 2023
- Snap Research Fellowship, 2022
- Meta PhD Fellowship Finalist, 2023
- School of Engineering Fellowship, Stanford University, 2019-2020
- HKSAR Government Targeted Scholarship (Full 4-year university scholarship)
- NUS Graduate Scholarship for ASEAN Nationals (Full masters scholarship)
- Google Women Techmakers Scholarship, 2016
- Epsilon Fund Award, HKUST Mathematics Department, 2017
- International Mathematical Olympiad (IMO) Bronze Medalist, 2012, 2013
- Philippine Mathematical Olympiad 1st runner-up, 2012, 2013
Research Intern, Project Starline, Google
Research Intern, Autodesk AI Lab
- Stanford CS 348n Guest Lecture, May 31, 2023, Neural Radiance Fields: Sparse View and Dynamic Scenes
- Stanford CS 348n Guest Lecture, May 24, 2023, Continuous and Discrete Shape Edits/Deformations
- Stanford CS 348n Guest Lecture, February 16, 2022, Neural Shape Variation and Generation
- Teaching Assistant, Winter 2021, Computer Graphics: Geometric Modeling/Processing (CS 348a), Stanford University
Interpretable & Actionable Models using Attribute & Uncertainty Information
- Deep-learning models can be difficult to understand and control intuitively due to the black-box nature of these models. However, such lack of interpretability and human actionability in the models’ decision processes make it difficult to trust these models in critical applications that affect the lives of people. We propose to alleviate these problems through the use of attribute and uncertainty models in deep networks.
HKUST Robotics Team, Remotely Operated Vehicle (ROV) Subteam
- Overall 3rd Place (Explorer Class) – 14th Annual MATE International Underwater Robotics Competition in St John’s, Newfoundland and Labrador, Canada
- Asia Champion in 2015 MATE Asia Regional Underwater Robotics Competition
- Built the main control software of the ROV, which operates with ROS and is controlled with an Xbox controller, and Qt GUI’s for the competition runs
- The team is composed of 15 engineers who built and designed the ROV from scratch.
Hobbies and Interests
For most of my pre-university life, I was into competitive mathematics, with geometry being a favorite topic. I competed in various math competitions both local and abroad representing the Philippine Team. During my spare time back at home, I now train elementary and high school students for international math competitions. I was part of the training team of the 2017-2020 PH IMO team, and I led the PH team to a number of elementary math competitions.
I also enjoy playing soccer, frisbee and scuba diving. I was part of the HKUST Women's Soccer Team back in my senior year.
Reviewer: CVPR, ICCV, ECCV, SIGGRAPH, SIGGRAPH Asia, BMVC, 3DV, AAAI, TVCG, Eurographics, Neurips, ICLR