Lin Shan
LS
Lin Shan
Presales Engineer @ DDN  ·  PhD Computer Science
AI infrastructure specialist and computer vision researcher. Currently driving AI/HPC storage solutions across APAC at DataDirect Networks. Former Research Fellow and team lead at ROSE Lab, NTU (2020–2024). PhD from University of Warwick, EU Marie Curie Fellow.
✉ Email LinkedIn Google Scholar YouTube Bilibili

About

I am currently a Presales Engineer at DataDirect Networks (DDN), specializing in AI infrastructure and high-performance storage solutions for the APAC region. Prior to this role, I was a Research Fellow at the Rapid-Rich Object Search (ROSE) Lab, Nanyang Technological University (NTU), where I led a 10-person team developing feasible AI solutions for real-world problems.

I obtained my Ph.D. in Computer Science from the University of Warwick (2015–2020), supervised by Prof. Chang-Tsun Li and Prof. Alex C. Kot. I was a Marie Sklodowska-Curie Fellow under the EU Horizon 2020 IDENTITY project, with research exchanges to Hong Kong Baptist University and NTU in 2016 and 2017. My research spans Person Re-identification, Deepfake Detection, and AI Domain Adaptation & Generalization.

Experience

Sep 2024 — Present
Presales Engineer
DataDirect Networks (DDN) · Singapore
Technical Presales and Solution Architecture for AI/HPC storage infrastructure across the APAC region.
May 2020 — Sep 2024
Research Fellow
ROSE Lab, Nanyang Technological University · Singapore
Led a 10-person research team focused on AI/ML, computer vision, and real-world deployment of intelligent surveillance systems.
Jul 2019 — Apr 2020
Project Officer
ROSE Lab, Nanyang Technological University · Singapore
AI and computer vision research projects.
Oct 2017 — Jul 2019
Research Assistant — Marie Curie Fellowship
ROSE Lab, Nanyang Technological University · Singapore
EU Horizon 2020 Marie Sklodowska-Curie Fellowship. Research on Person Re-identification and cross-domain adaptation.
Dec 2016 — Mar 2017
Senior Research Assistant — Marie Curie Fellowship
Hong Kong Baptist University · Hong Kong
Research exchange under EU Horizon 2020 IDENTITY project.
Mar 2015 — Apr 2015
Engineering Intern
Tencent / Capcom · Shanghai
Monster Hunter Online game development.
Jul 2014 — Aug 2014
Engineering Intern
ZTE / Boom-T · Guangzhou
Power grid communication systems.

Education

Oct 2015 — Apr 2020
Doctor of Philosophy (Ph.D.) — Computer Science
University of Warwick · United Kingdom
Supervised by Prof. Chang-Tsun Li and Prof. Alex C. Kot. Research: Person Re-identification, Domain Adaptation/Generalization. EU Marie Sklodowska-Curie Fellow.
2012 — 2015
Bachelor of Science (B.Sc.) — Computer Science
University of Warwick · United Kingdom

Projects

DSTA-ROSE: AI for Human Re-Identification Across Cameras
Defence Science and Technology Agency (DSTA)
AI-powered person re-identification system for trajectory tracking and person matching across camera networks. Deployed for COVID-19 contact tracing at Changi Exhibition Centre isolation facility.
[Trajectory Tracking] [Person Matching] Read More →
KLASS-ROSE: Image Enhancement for Person Re-ID in Low Light
KLASS
Image enhancement techniques and domain adaptation methods to improve person re-identification under low-light and challenging lighting conditions.
DSTA-ROSE: Attire Insensitive Person Re-ID
Defence Science and Technology Agency (DSTA)
Robust person re-identification that remains effective when subjects change clothing, addressing a critical challenge for long-term surveillance applications.

Competitions

🏆
1st
IEEE SPS 5-Minute Video Clip Contest — “Fight the Pandemic”
Visual Analytic System for Pandemic Management During COVID-19
[YouTube] [News] Read More →
🏈
2nd RU
NUS-MIT Healthcare AI Datathon 2020
Domain Shift Problem in Chest X-ray Diagnosis AI Model — 2nd Runner-up
[YouTube] [News] Read More →
Top 10
9th
Trusted Media Challenge 2021 — Deepfake Detection
9th Place Overall, 2nd among Universities — multi-modal deepfake and lip-sync detection
[Deepfake Demo] [Lip Sync Demo] Read More →

Datasets

NTU Outdoor Dataset
Location: Nanyang Technological University Outdoor Campus
Cameras: 51 real-world surveillance cameras (38 in released version)
Identities: 805 IDs (549 in released version)
Attributes: 40 soft biometric attributes
Coverage: 24-hour images (day & night)
Download Dataset →   Full Details →

Publications

A Unified Deep Semantic Expansion Framework For Domain-Generalized Person Re-Identification
Neurocomputing · 2024
[arXiv]
Color Space Learning for Cross-Color Person Re-Identification
ICME · 2024
[arXiv]
Diverse Deep Feature Ensemble Learning for Omni-Domain Generalized Person Re-identification
ICMIP · 2024
[ACM DL]
Aligned Divergent Pathways for Omni-Domain Generalized Person Re-Identification
ICEEE · 2024
[arXiv]
Adversarial Pairwise Reverse Attention for Camera Performance Imbalance in Person Re-Identification: New Dataset and Metrics
ICIP 2022 · Eugene P.W. Ang, Shan Lin, Alex C Kot
[arXiv]
DEX: Domain Embedding Expansion for Generalized Person Re-identification
BMVC 2021 · Eugene P.W. Ang, Shan Lin, Alex C Kot
[BMVC] [arXiv] [Video]
Multi-Domain Adversarial Feature Generalization for Person Re-Identification
IEEE Transactions on Image Processing (TIP) · 2020 · Shan Lin, Chang-Tsun Li, Alex C Kot
[IEEE] [arXiv]
Person Re-identification with Soft Biometrics through Deep Learning
Deep Biometrics, Springer · 2020 · Shan Lin, Chang-Tsun Li
[Springer]
Scalable Deep Feature Learning for Person Re-Identification
PhD Thesis · University of Warwick · 2019
[Warwick]
Homogeneous Feature Transfer and Heterogeneous Location Fine-tuning for Cross-City Property Appraisal Framework
AusDM 2018, Springer · Yihan Guo, Shan Lin, Xiao Ma, Jay Bal, Chang-Tsun Li
[Springer] [arXiv]
Multi-task Mid-level Feature Alignment Network for Unsupervised Cross-Dataset Person Re-Identification
BMVC 2018 Spotlight · Shan Lin, Haoliang Li, Chang-Tsun Li, Alex C Kot
[BMVC] [arXiv] [YouTube]
End-to-End Correspondence and Relationship Learning of Mid-Level Deep Features for Person Re-Identification
DICTA 2017 · Shan Lin, Chang-Tsun Li
[IEEE]