Introduction

In recent years, foundation models have emerged as transformative tools in computer vision, offering powerful zero-shot and few-shot learning capabilities across a wide range of tasks. Their integration into visual anomaly detection—a critical and high-stakes field spanning healthcare, industrial inspection, security, and autonomous systems—has opened new frontiers in both research and real-world applications. This tutorial aims to deliver a comprehensive and timely overview of the role of foundation models in visual anomaly detection. We will cover multiple visual modalities, including 2D images, 3D images, and videos—each presenting unique challenges and necessitating modality-specific solutions. Specifically, we will delve into the entire pipeline, from data (pre-)training and prompt engineering to methodological innovations, inference strategies, and deployment in real-world environments. Key topics include zero- and few-shot learning, pseudo-labeling, anomaly generation, and multi-modal alignment between vision and language. To facilitate a deep and practical understanding of these areas, the tutorial will bring together leading experts from both academia and industry. Through in-depth technical presentations and discussions, participants will gain valuable insights into the latest advances, real-world applications, and open challenges shaping this rapidly evolving field.

Course Description

This tutorial will be held as a half-day event on 19 October 2025, from 9:00 AM to 12:00 PM (PDT) in room 307 A. 

The planned agenda is as follows:

9:00-9:15

[15 mins]  Opening Remarks  [Guansong Pang] [slides]

9:15-10:00

[45 mins presentation & QA] FM-driven Image Anomaly Detection  [Jiawen Zhu, Guansong Pang] [slides]

10:00-10:45

[45 mins presentation & QA] Anomaly Generation with FMs   [Ismail Nejjar, Olga Fink] [slides]

10:45-11:30

[45 mins presentation & QA] Benchmarking Realistic Anomaly Detection  [Jinlong Peng, Chenjie Wang] [slides]

11:30-12:15

[45 mins presentation & QA] FM-driven Video Anomaly Detection and Reasoning  [Peng Wu] [slides]

Organizers

Guest Speakers

Jiawen Zhu, PhD Candidate

Singapore Management University (SMU), Singapore (jwzhu.2022@phdcs.smu.ed.sg) 

Jiawen Zhu is currently a fourth-year PhD student at Singapore Management University and supervised by Prof. Guansong Pang. She has published and presented multiple papers at top venues, including CVPR and ICCV. Her research interests lie in the field of computer vision and open world learning, particularly in foundation models for anomaly detection. 

Dr. Peng Wu, Associate Professor 

Northwestern Polytechnical University, Xi’an, China (pengwu@nwpu.edu.cn) 

Peng Wu is currently a tenure-track associate professor at Northwestern Polytechnical University. He received his Ph.D. degree and B.Eng from Xidian University in 2022 and 2017. He was a Research Intern of Alibaba DAMO Academy and was a Program Committee Member of VAND 2.0/3.0@CVPR workshops. His research interests include video anomaly detection and video retrieval. He has published more than 40 papers in top conferences and journals, such as CVPR, ECCV, ACM MM, AAAI,and TIP. 

Dr. Olga Fink, Assistant Professor of Intelligent Maintenance and Operations Systems 

École polytechnique fédérale de Lausanne (EPFL), Lausanne, Switzerland (olga.fink@epfl.ch) 

Olga Fink is an assistant professor leading the Laboratory of Intelligent Maintenance and Operations Systems at EPFL since March 2022. She has received numerous awards and honors including being named one of Switzerland’s “Top 100 Women in Business” in 2018, recognition as a Young Scientist by the World Economic Forum and World Laureate Forum, and being named a Fellow of the Prognostics and Health Management Society in 2023. Her research focuses on advancing intelligent maintenance and operations of complex infrastructure and industrial assets through the development of Physics-Informed Machine Learning, Graph Neural Networks, Domain Adaptation, Self-Supervised Learning, Deep Reinforcement Learning, and Multi-Agent Systems. 

Dr. Chengjie Wang

Shanghai, China (chengjiewang@sjtu.edu.com) 

Chengjie Wang has published more than 100 papers on major computer vision and artificial intelligence conferences, such as TPAMI, TIP, IJCV, CVPR, ICCV, and NeurIPS, including 20+ Industrial Anomaly Detection related papers. His research interests include computer vision and machine learning. He holds more than 100 patents in these areas. 

Dr. Guansong Pang, Assistant Professor of Computer Science 

Singapore Management University (SMU), Singapore (gspang@smu.edu.sg) 

Guansong Pang obtained his PhD degree at University of Technology Sydney in 2019. His research interest lies in machine learning and computer vision. He has published more than 80 papers in top conferences and journals, such as CVPR, ICCV, ECCV, NeurIPS, ICLR, ACM MM, KDD, AAAI, and IJCAI. He is the main speaker of KDD’21, WSDM’21 and CVPR’23 tutorials on deep anomaly detection, the main organizer of ANDEA, AI4AN and VAND workshop series on anomaly and novelty detection at IJCAI, KDD and CVPR, a lead guest editor of IEEE TNNLS on “deep learning for anomaly detection”, an Area Chair of CVPR, NeurIPS, ICLR, and KDD and Associate Editor of IEEE TNNLS and Elsevier PRJ handling anomaly detection and open world learning submissions.

Dr. Jinlong Peng

Shanghai, China (pjl_1995@163.com)

Jinlong Peng has published more than 30 papers on major computer vision and artificial intelligence conferences, such as TPAMI, CVPR, ICCV, ECCV, and NeurIPS. His research interests include computer vision and machine learning. He holds more than 30 patents in these areas.

Contact

Jiawen Zhu, jwzhu.2022@phdcs.smu.ed.sg

Guansong Pang, gspang@smu.edu.sg