My work lies at the intersection of deep learning, computer vision, and robotics.
Work
Tech lead of the Perception team. Working on our end-to-end driving model.
TLM of the Vehicle Perception team, leading a team of 10+ engineers.
❖ Foundation Sensor Model
Led the cross-functional effort to build and ship the first end-to-end perception model at Woven. Owned multiple core perception capabilities: 3D detection-tracking, sensor fusion, and large-scale foundation training.
❖ ML Edge Compute
Built the ML edge deployment capability from the ground up. Led a cross-functional team to compile, optimize, and deploy our ML models on both Qualcomm and NVIDIA hardware platforms. Owned the ML deployment stack: model conversion, performance optimization, and edge evaluation.
Lyft Level 5 was acquired by Woven by Toyota in July 2021.
High-fidelity 3D eye segmentation using implicit neural representations.
Lidar-based detection model for large vehicles.
Selected Publications
LCD: Learned Cross-domain Descriptors for 2D-3D Matching
Quang-Hieu Pham, Mikaela Angelina Uy, Binh-Son Hua, Duc Thanh Nguyen, Gemma Roig, and Sai-Kit Yeung
In AAAI, 2020.
A*3D: An Autonomous Driving Dataset in Challenging Environment
Quang-Hieu Pham*, Pierre Sevestre*, Ramanpreet Singh Pahwa, Huijing Zhan, Chun Ho Pang, Yuda Chen, Armin Mustafa, Vijay Chandrasekhar, and Jie Lin
In ICRA, 2020.
JSIS3D: Joint Semantic-Instance Segmentation of 3D Point Clouds
Quang-Hieu Pham, Duc Thanh Nguyen, Binh-Son Hua, Gemma Roig, and Sai-Kit Yeung
In CVPR, 2019.
Real-time Progressive 3D Semantic Segmentation for Indoor Scenes
Quang-Hieu Pham, Binh-Son Hua, Duc Thanh Nguyen, and Sai-Kit Yeung
In WACV, 2019.
SceneNN: A Scene Meshes Dataset with aNNotations
Binh-Son Hua, Quang-Hieu Pham, Minh-Khoi Tran, Duc Thanh Nguyen, Lap-Fai Yu, and Sai-Kit Yeung
In 3DV, 2016.