Portfolio item number 1
TL;DR:
Posts by Collection
portfolio
Portfolio item number 2
TL;DR:
publications
Robustness of trajectory prediction models under map-based attacks
Zhihao Zheng, Xiaowen Ying, Zhen Yao, Mooi Choo Chuah
TL;DR: We study the robustness of trajectory prediction models with image-based or node-based map representations under our newly proposed map-based adversarial attacks and propose two effective defense mechanisms to defend against such map-based attacks.
[Paper] | [BibTeX]
Goal-lbp: Goal-based local behavior guided trajectory prediction for autonomous driving
Zhen Yao, Xin Li, Bo Lang, Mooi Choo Chuah
TL;DR: We propose a goal-based local behavior guided model, Goal-LBP, using historical paths at a certain location (referred as local behavior data) to generate potential goals and guide the prediction of trajectories conditioned on such goals.
[Paper] | [BibTeX]
CrackNex: a Few-shot Low-light Crack Segmentation Model Based on Retinex Theory for UAV Inspections
Zhen Yao, Jiawei Xu, Shuhang Hou, Mooi Choo Chuah
TL;DR: We propose CrackNex, a framework that utilizes reflectance based on Retinex Theory and few-shot strategy to address limited low-light crack training data issue, and present the first benchmark dataset, LCSD, for low-light crack segmentation.
[Paper] | [Code] | [arXiv] | [BibTeX]
Event-guided Low-light Video Semantic Segmentation
Zhen Yao, Mooi Choo Chuah
TL;DR: We propose EVSNet, a lightweight framework that introduces event modality to guide the learning of a unified illumination-invariant representation and it outperforms SOTA methods with up to 11× higher parameter efficiency.
[Paper] | [Code] | [arXiv] | [BibTeX]
Learning Flow-Guided Registration for RGB-Event Semantic Segmentation
Zhen Yao, Xiaowen Ying, Zhiyu Zhu, Mooi Choo Chuah
TL;DR: We recast RGB-Event segmentation from fusion to registration via a novel flow-guided registration-centric framework and introduce a new Motion-enhanced Event Tensor (MET) representation, mitigating spatiotemporal and modal misalignments.
[Paper] | [Code] | [arXiv] | [BibTeX]
talks
Talk 1 on Relevant Topic in Your Field
TL;DR:
Tutorial 1 on Relevant Topic in Your Field
TL;DR:
Talk 2 on Relevant Topic in Your Field
TL;DR:
Conference Proceeding talk 3 on Relevant Topic in Your Field
TL;DR: