|
Christina Dao Wen Lee I'm a PhD Candidate and Graduate Tutor at National University of Singapore (Mechanical Engineering) and Advanced Robotics Centre (ARC). I am supervised by Prof Marcelo Ang and Dr Zhang Hongying. Research Keywords: Sensor Fusion, Robot Vision, Machine Learning, Visual Place Recognition, Object Detection, Segmentation Scene Understanding, Radar |
|
Research
I'm interested in multi-sensor fusion for object detection and visual place recognition. I also research on techniques for optimising performance and speed for real-world implementations on various platforms. Representative paper(s) are highlighted.
|
|
DINO-CoDT: Multi-class Collaborative Detection and Tracking with Vision Foundation Models Xunjie He* ,Christina Dao Wen Lee* , Meiling Wang, Chengran Yuan, Zefan Huang, Yufeng Yue†, Marcelo H Ang Jr Under Review paper DINO-CoDT, a novel framework for collaborative detection and tracking in complex multi-class autonomous driving scenarios.Through extensive experiments on V2X-Real and OPV2V benchmark, DINO-CoDT demonstrates significant improvements in both detection and tracking accuracy. |
|
DINO-MOT: 3D Multi-Object Tracking With Visual Foundation Model for Pedestrian Re-Identification Using Visual Memory Mechanism Min Young Lee†, Christina Dao Wen Lee, Jianghao Li, and Marcelo H. Ang Jr RAL 2024 , ICRA 2025 paper
|
|
ARC-BEV: Attentive Radar-Camera fusion 3D object detection in Bird-Eye-View space for autonomous driving Lyuyu Shen†, Jianghao Li, Christina Dao Wen Lee, Min Young Lee, Andreas Hartmannsgruber and Marcelo H. Ang Jr Presented at ISER, 2023
In this paper, we propose a straightforward and efficient fusion framework for camera and radar in BEV space.
|
|
Hot-NetVLAD: Learning discriminatory key points for visual place recognition Zhikai Li, Christina Dao Wen Lee†, Beatrix Xue Lin Tung, Zefan Huang, Daniela Rus, Marcelo H Ang RAL, 2023 paper Hot-NetVLAD implements a hot-spot detector on a learned local key-patch descriptor algorithm for Visual Place Recognition (VPR), thereby greatly cutting down the size of features extracted. Furthermore, identified hot-spots bring new insights to key regions required for VPR, as they tend to fall on distinguishable static objects in the scene. |
|
Online obstacle trajectory prediction for autonomous buses Yue Linn Chong*, Christina Dao Wen Lee*†, Liushifeng Chen, Chongjiang Shen, Ken Kok Hoe Chan, Marcelo H Ang Jr Machines, 2022 Feature Paper SGAB Dataset / paper
In this paper, we present the development of a modular pipeline for the long-term prediction of dynamic obstacles’ trajectories for an autonomous bus. Our Singapore autonomous bus (SGAB) dataset evaluated the pipeline’s performance. The dataset is publicly available online.
|
Academic ServiceReviewer Activities
· Robotics: Science and Systems (RSS) Volunteer· IEEE RAS Technical Committee Computer & Robot Vision, Student Representative· International Symposium on Experimental Robotics (ISER) 2023 - Admin |
