Haoda Li

Software Engineer in Graphics and Machine Learning

About

I'm skillful and passionate in developing new innovations and real-world solutions for parallel computing applications in machine learning and 3D graphics. Currently, I'm a ML compiler engineer who works on accelerating and optimizing GenAI models on cutting-edge ML chips. Previously, I completed my Master of Engineering in EECS at UC Berkeley, where I worked on large-scale scene reconstructions, supervised by Prof. Avideh Zakhor, and I obtained my Bachelor of Science degree at U of Toronto, with a focus on artificial intelligence and computer vision.

Resume/CV

Education

Master of Engineering
Electrical Engineering and Computer Sciences (GPA 3.93/4.00)
University of California, Berkeley
2022-2023
Honours Bachelor of Science
Computer Science and Data Science (GPA 3.91/4.00)
University of Toronto
2017-2022

Featured Experiences

Graduate Research Student

Supervised by Avideh Zakhor
VIP Lab, UC Berkeley

Aug. 2022 - May. 2023

Working on end-to-end drone systems for high quality indoor 3D reconstruction and measurements.
Integrating neural rendering technology into large scale 3D mapping and reconstruction.
Research Engineer

Supervised by Juwei Lu
Noah's Ark Lab, Huawei Canada Inc.

May. 2020 - Aug. 2021

Working on a novel method for event localization and classification in videos.
Assisting research on self-supervised video indexing and retrieval.
Researching and integrating video understanding methods for video editing applications.
Assisting research on hand tracking and human action recognition.
Research Student

Supervised by Animesh Garg
PAIR Lab, University of Toronto

Aug. 2021 - Jul. 2022

Wokring on a novel method for assembling aribitrary shaped 3D objects without additional guidiance.
Creating simulation environment for generating fractured 3D objects.

Publications

Scalable MAV Indoor Reconstruction with Neural Implicit Surfaces

Haoda Li , Puyuan Yi , Yunhao Liu , Avideh Zakhor
ICCV 2023 Workshop on Computer Vision Aided Architectural Design

Paper

SPIDR: SDF-based Neural Point Fields for Illumination and Deformation

Ruofan Liang , Jiahao Zhang , Haoda Li , Chen Yang , Nandita Vijaykumar
CVPR 2023 Workshop on Advances in NeRF for the Metaverse

Paper Project Code

Neural Shape Mating Self-Supervised Object Assembly with Adversarial Shape Priors

Yun-Chun Chen , Haoda Li , Dylan Turpin , Alec Jacobson , Animesh Garg
Computer Vision and Pattern Recognition (CVPR), 2022

Paper Project Code

Decompose the Sounds and Pixels, Recompose the Events

Varshanth Rao , Md Ibrahim Khalil , Haoda Li , Peng Dai , Juwei Lu
Confernce on Artificial Intelligence (AAAI), 2022

Paper Project

Dual Perspective Network for Audio Visual Event Localization

Varshanth Rao , Md Ibrahim Khalil , Haoda Li , Peng Dai , Juwei Lu
European Conference on Computer Vision (ECCV), 2022

Paper