
Cody Reading
I am a 3D computer vision researcher, interested in perception, reconstruction, and generation. I aim to develop tools that understand our surrounding 3D world.
I received my master's degree from the University of Toronto supervised by Steven Waslander, and received my bachelor's degree from the University of Waterloo. I worked at MARZ developing Vanity AI, an AI solution for facial editing for VFX, and at Huawei Noah’s Ark Lab working on indoor mobile robotic applications.
Experience
Senior Researcher — Huawei Noah's Ark Lab
Built indoor mobile robotic pipelines to enable manipulation and navigation, involving implementation of methods in 3D scene graph estimation, segmentation, SLAM, path planning, and control.
3D Content Creation Researcher — Simon Fraser University
Implemented novel techniques in 3D content creation and generative models, involving optimizing NeRF and 3D Gaussian representations with diffusion guidance.
Machine Learning Research Associate — MARZ
Developed a facial de-aging tool Vanity AI designed for VFX applications, achieving 300x speed up compared to traditional VFX workflows.
3D Perception Researcher — University of Toronto
Developed methodologies in autonomous vehicle 3D perception, achieving 1st and 2nd place on 3D monocular object detection and 3D multi-object tracking benchmarks respectively.
Software Engineer (Autonomous Driving) — NVIDIA Corporation
Integrated a vehicle trajectory generation library within the NVIDIA DriveWorks SDK using C++ to generate a sequence of vehicle poses from GPS, IMU, and CAN sensor data.
Semantic Segmentation Research Co-op — University of Waterloo
Developed semantic segmentation training infrastructure to support unified training of the SegNet and FCN methods on the Cityscapes, Playing-for-data, and Synthia datasets.
Publications

CVPR 2024
Highlight
Bayes' Rays: Uncertainty Quantification for Neural Radiance Fields
Lily Goli, Cody Reading, Silvia Sellán, Alec Jacobson, Andrea Tagliasacchi
A post-hoc framework to evaluate uncertainty in any pre-trained NeRF without modifying the training process.

CVPR 2024
BANF: Band-limited Neural Fields for Levels of Detail Reconstruction
Ahan Shabanov, Shrisudhan Govindarajan, Cody Reading, Lily Goli, Daniel Rebain, Kwang Moo Yi, Andrea Tagliasacchi
A method for band-limited frequency decomposition in neural fields

CVPR 2021
Oral Presentation
Categorical Depth Distribution Network for Monocular 3D Object Detection
Cody Reading, Ali Harakeh, Julia Chae, Steven Waslander
Estimating categorical depth distributions results in accurate image feature projection into 3D

ITSC 2018
Unlimited Road-scene Synthetic Annotation (URSA) Dataset
Matt Angus, Mohamed ElBalkini, Samin Khan, Ali Harakeh, Oles Andrienko, Cody Reading, Steven Waslander, Krzysztof Czarnecki
Synthetic dataset generation for autonomous vehicle semantic segmentation based on GTA V
Education
University of Toronto
M.A.Sc. in Aerospace Engineering
Advisor: Prof. Steven Waslander
University of Waterloo
B.A.Sc. in Mechatronics Engineering