Emily Steiner
Stanford University TomKat Center University of Waterloo

I am a PhD candidate at Stanford University, advised by Professor Iro Armeni in the Gradient Spaces Lab. My research focuses on deep learning-based computer vision for understanding evolving 3D scenes over time (4D), spanning both dynamically changing lived-in spaces (robotics and embodied AI) and structurally evolving scenes with large-scale spatio-temporal changes (building construction monitoring). My research is graciously funded by the TomKat Center Graduate Fellowship for Translational Research.

Previously, I completed my BASc in Mechatronics Engineering with a Computing Option at the University of Waterloo, graduating with distinction. I gained industry experience through co-op positions in hardware research and software engineering.



CONTACT

Email: easteine [at] stanford [dot] edu

github link   linkedin link   CV



Publications

My current research focuses on 3D/4D scene understanding, spatio-temporal semantic representations, and datasets, benchmarks, and metrics for dynamically changing scenes. If you find my work interesting and would like to chat, feel free to reach out!



ReScene4D

ReScene4D: Temporally Consistent Semantic Instance Segmentation of Evolving Indoor 3D Scenes
Emily Steiner, Jianhao Zheng, Henry Howard-Jenkins, Chris Xie, Iro Armeni
• Under Review, 2026
• Code / Project / Preprint (coming soon)



Research Positions

Gradient Spaces Lab | PhD Candidate | Advisor: Iro Armeni, Stanford University
April 2024 - present | Lab website ↗


Computational Imaging Lab | Rotation PhD Student | Advisor: Gordon Wetzstein, Stanford University
Winter 2024 | Lab website ↗



Projects



Industry Experience

May - August 2022

Hardware Research & Development Engineering Co-op

Lumafield

Investigated CT scanner imaging limitations to characterize the trade-off between sharpness and scan time. Determined system parameter improvements for a 3x speed increase with no loss in quality; changes implemented on customer machines and showcased at trade shows. Created an exploratory multi-detector prototype, developed an image processing pipeline using OpenCV in Python, and implemented a joint iterative calibration algorithm based on current literature. Improved capture speed by 200% with minimal impact on image quality.

January - April 2022

Product Development Intern (Electrical / Mechatronics)

Inertia Product Development

Designed and implemented a LiDAR point cloud visualizer using Python and Qt to interface with a Robot Operating System (ROS) backend, enabling real-time visualization of sensor data for autonomous systems development. Conducted feasibility studies for customer projects, using rapid prototyping to evaluate technical approaches and determine optimal solutions for product development success.

May - August 2021

Aerospace Engineer

Canadensys

Designed embedded state machine architecture and PID controller, implementing firmware on an STM32 Microcontroller in C for a high precision BLDC motor controller. Performed control system and stability analysis using Simulink. Conducted root cause analysis to identify a fundamental hardware design error causing unstable motor feedback. Implemented a corrective software algorithm for Hall Effect sensor processing, salvaging the PCB design and avoiding costly hardware redesign.

January - April & September - December 2020

Software Engineering Co-op Student

ExactEarth (Spire)

Developed autonomous operations management software responsible for communication procedures, telemetry collection, and recovery processes of the EV10 satellite. Supported the software control system throughout satellite commissioning and post-launch operations (launched Sept 2020).



Teaching & Mentorship



Leadership & Involvement



Honours & Awards