hi, i’m kathryn!
i’m currently at google, working on beam (formerly project starline).
my work is on fast, generalizable novel view synthesis.
before this, I was at google x, harvard, and ucla.
recent news:
- Our paper LVT: Large-Scale Scene Reconstruction via Local View Transformers was accepted to SIGGRAPH Asia 2025, and the Women in Machine Learning Workshop @ NeurIPS 2025.
- Our paper Quark: Real-time, High-resolution, and General Neural View Synthesis won a Best Paper award at SIGGRAPH Asia 2024.
- I submitted my fifth <unreleased> patent at Google.
about me
academia
&
industry
industry research
2021-present
Google, Project Starline
we develop and train novel view synthesis models for telepresence
ai residency
2020-2021
Google X
we worked on ml for an industrial inverse design problem
my dissertation
2016-2021
Harvard SEAS
we turned shape from shading into algebraic geometry
graduate research
2014-2021
Harvard SEAS
we developed resource allocation algorithms
undergraduate research
2009-2014
UCLA Mathematics
i studied wavelets of customizable support
more about me
coding
python (JAX, FLAX, NumPy, TensorFlow)
mathematica, magma, MATLAB
blender, illustrator, photoshop
licenses & awards
hipaa – human subjects (2014)
certificate of distinction in teaching (2016)
service
reviewer for CVPR 2025
reviewer for ICCV 2025
what people are saying
she likes small animals
everyone
she’s under 5’4
everyone
blog
a few thoughts
- Advice to Applied Math Graduate StudentsI graduated in person this weekend, and in reflecting on my doctoral journey I’d like to share some strategies that helped me as an applied math grad student. Maintain research logs for yourself every day. By logs, I mean mini progress reports. And write them in LaTeX so you can copy and paste snippets ofContinue reading “Advice to Applied Math Graduate Students”
- Protected: Peripheral Nerve Stimulation in MRI: a Patient’s PerspectiveThis content is password protected.
- DissertationI am thrilled to share my dissertation, A Lighting-Invariant Approach to Local Shape from Shading. Citation Heal, Kathryn. 2021. A Lighting-Invariant Approach to Local Shape from Shading. Doctoral dissertation, Harvard University Graduate School of Arts and Sciences. Abstract Shape from shading is a classical problem in computer vision, in which the depth field of anContinue reading “Dissertation”
kathryn heal, ph.d. 