Muhammad Aqeel

Open to collaborations and positions

Postdoctoral Research Fellow
Department of Engineering for Innovation Medicine · University of Verona, Italy

Anomaly detection·Unsupervised learning·Vision-language models·Generative models

Download CV

My research addresses visual anomaly detection in industrial settings where defective samples are scarce or unavailable. I work on unsupervised anomaly detection, where only normal samples are available for training, and on zero-shot and generalist anomaly detection, where anomalies must be detected in object categories not seen during training. I approach these problems from three directions: confident pseudo-labeling, which stabilizes self-training under distribution shift, allowing the detector to refine what it treats as normal without manual relabeling; vision-language adapters, which adapt foundation-model representations to learn a category-agnostic notion of normality that generalizes across object categories; and diffusion-based defect synthesis, which leverages pre-trained diffusion models to generate realistic anomalous samples for data augmentation when real defects cannot be collected.

A second line extends these methods to medical image analysis, where abnormal samples are equally scarce and missed detections carry higher cost. In this domain, I use text-guided diffusion augmentation and representation transfer to detect abnormalities in biomedical images despite limited annotated data.

I completed my PhD at the University of Verona with Francesco Setti and Marco Cristani, and my Master's at the Beijing University of Posts and Telecommunications with Xiaohong Liu.

News

Jun 2026 LiZAD accepted at IEEE COINS 2026.
Apr 2026 AVA-DINO accepted at ICIP 2026.
Mar 2026 ClipSum accepted at ICPR 2026.
Mar 2026 Started as Postdoctoral Research Fellow at the University of Verona.
Feb 2026 Successfully defended my PhD at the University of Verona.

See all news →

Selected publications

CoMet

Towards Real Unsupervised Anomaly Detection Via Confident Meta-Learning

Muhammad Aqeel, Shakiba Sharifi, Marco Cristani, Francesco Setti

ICCV 2025·PDF·Code

AVA-DINO

Anomaly-Aware Vision-Language Adapters for Zero-Shot Anomaly Detection

Muhammad Aqeel, Maham Nazir, Uzair Khan, Marco Cristani, Francesco Setti

ICIP 2026·PDF·Code

ClipSum

Multimodal Abstractive Summarization of Instructional Videos with Vision-Language Models

Maham Nazir, Richong Zhang, Muhammad Aqeel, Francesco Setti

ICPR 2026·PDF·Code

RAD

Robust Anomaly Detection in Industrial Environments via Meta-Learning

Muhammad Aqeel, Shakiba Sharifi, Marco Cristani, Francesco Setti

ICCV 2025 Workshops (Oral)·PDF

Semantic Parsing

Semantic Parsing for Aspect-Based Sentiment Analysis

Muhammad Aqeel, Francesco Setti

IEEE Access 2025·PDF

See all publications →

Publications

See also Google Scholar for citation metrics.

2026

[15]

LiZAD: A Lightweight Zero-Shot Anomaly Detection Framework for Industrial Manufacturing

Uzair Khan, Luigi Capogrosso, Muhammad Aqeel, Francesco Setti, Michele Magno, Marco Cristani

IEEE COINS 2026 PDF·Code
[14]

Anomaly-Aware Vision-Language Adapters for Zero-Shot Anomaly Detection

Muhammad Aqeel, Maham Nazir, Uzair Khan, Marco Cristani, Francesco Setti

ICIP 2026 PDF·Code
[13]

Multimodal Abstractive Summarization of Instructional Videos with Vision-Language Models

Maham Nazir, Richong Zhang, Muhammad Aqeel, Francesco Setti

ICPR 2026 PDF·Code

2025

[12]

Semantic Parsing for Aspect-Based Sentiment Analysis

Muhammad Aqeel, Francesco Setti

IEEE Access 2025 PDF
[11]

Towards Real Unsupervised Anomaly Detection Via Confident Meta-Learning

Muhammad Aqeel, Shakiba Sharifi, Marco Cristani, Francesco Setti

ICCV 2025 PDF·Code
[10]

A Contrastive Learning-Guided Confident Meta-learning for Zero Shot Anomaly Detection

Muhammad Aqeel, Danijel Skocaj, Marco Cristani, Francesco Setti

ICCV 2025 Workshops Oral PDF
[9]

Robust Anomaly Detection in Industrial Environments via Meta-Learning

Muhammad Aqeel, Shakiba Sharifi, Marco Cristani, Francesco Setti

ICCV 2025 Workshops Oral PDF
[8]

Diffusion-Based Data Augmentation for Medical Image Segmentation

Maham Nazir, Muhammad Aqeel, Francesco Setti

ICCV 2025 Workshops PDF
[7]

RoadFusion: Latent Diffusion Model for Pavement Defect Detection

Muhammad Aqeel, Kidus Dagnaw Bellete, Francesco Setti

ICIAP 2025 Oral PDF
[6]

ExDD: Explicit Dual Distribution Learning for Surface Defect Detection via Diffusion Synthesis

Muhammad Aqeel, Federico Leonardi, Francesco Setti

ICIAP 2025 PDF·Code
[5]

Latent Space Synergy: Text-Guided Data Augmentation for Direct Diffusion Biomedical Segmentation

Muhammad Aqeel, Maham Nazir, Zanxi Ruan, Francesco Setti

ICIAP 2025 Workshops PDF
[4]

Self-Supervised Iterative Refinement for Anomaly Detection in Industrial Quality Control

Muhammad Aqeel, Shakiba Sharifi, Marco Cristani, Francesco Setti

VISAPP 2025 Oral PDF

2024

[3]

Meta Learning-Driven Iterative Refinement for Robust Anomaly Detection in Industrial Inspection

Muhammad Aqeel, Shakiba Sharifi, Marco Cristani, Francesco Setti

ECCV 2024 Workshops Oral PDF
[2]

Self-supervised Learning for Robust Surface Defect Detection

Muhammad Aqeel, Shakiba Sharifi, Marco Cristani, Francesco Setti

DeLTA 2024 Oral PDF
[1]

Diffusion-based Image Generation for In-distribution Data Augmentation in Surface Defect Detection

Luigi Capogrosso, Federico Girella, Francesco Taioli, Michele Dalla Chiara, Muhammad Aqeel, Franco Fummi, Francesco Setti, Marco Cristani

VISAPP 2024 PDF·Code

News

A timeline of recent updates, paper acceptances, talks, and milestones.

2026

Jun 2026 LiZAD accepted at IEEE COINS 2026. A lightweight zero-shot anomaly detection framework for industrial manufacturing.
Apr 2026 AVA-DINO accepted at ICIP 2026. Anomaly-aware vision-language adapters for zero-shot anomaly detection.
Mar 2026 ClipSum accepted at ICPR 2026. Multimodal abstractive summarization of instructional videos.
Mar 2026 Started as Postdoctoral Research Fellow at the University of Verona.
Feb 2026 Successfully defended my PhD at the University of Verona.

2025

Jul 2025 CoMet accepted at ICCV 2025; two more papers accepted at ICCV 2025 Workshops (both Oral).
Jun 2025 Started a 6-month research stay at the University of Ljubljana, Slovenia.

2024

Sep 2024 Paper accepted at ECCV 2024 Workshops (Oral).
Jul 2024 Paper accepted at DeLTA 2024 (Oral).

2022

Dec 2022 Started PhD in Computer Science at the University of Verona, Italy.

2019 - 2021

Jun 2021 Graduated with Master's in Computer Science from BUPT, Beijing.
Sep 2019 Started Master's in Computer Science at Beijing University of Posts and Telecommunications, China (CSC scholarship).

Grants

Funded research projects and scholarships.

2022 - 2025

PNRR PhD Scholarship

Italian Ministry of University and Research, via University of Verona

Role: PhD Fellow · National Recovery and Resilience Plan funding for doctoral research on industrial anomaly detection.

June 2025 - November 2025

Mobility Research Grant

University of Ljubljana, Slovenia

Role: Visiting Researcher · Six-month research stay collaborating with Prof. Danijel Skočaj's group on zero-shot anomaly detection.

2019 - 2021

Chinese Government Scholarship (CSC)

China Scholarship Council, via Beijing University of Posts and Telecommunications

Role: Scholarship Recipient · Supported MSc research on natural language processing.

Career

Education

2022 - 2026

Ph.D. in Computer SciencePhD

University of Verona, Italy

Supervisors: Prof. Francesco Setti & Prof. Marco Cristani · Anomaly detection for industrial manufacturing

2019 - 2021

M.Sc. in Computer ScienceMSc

Beijing University of Posts and Telecommunications, China

Supervisor: Prof. Xiaohong Liu · Natural language processing and sentiment analysis

2013 - 2017

Experience

2026 - present

Postdoctoral Research FellowPostdoc

University of Verona, Italy

June 2025 - November 2025

Visiting ResearcherVisiting

University of Ljubljana, Slovenia

2023 - 2025

Teaching AssistantTeaching

University of Verona, Italy

Computer Vision and Machine Learning, Department of Engineering for Innovation Medicine

2021 - 2022

Research AssistantResearch

Huazhong University of Science and Technology, Wuhan, China

2017 - 2019

Industry RolesIndustry

Pakistan

Software Quality Assurance Engineer at UIIT, Arid Agriculture University · Earlier, Software Developer at eConceptions (Pvt) Ltd.

Awards & honours

2025

Certificate of Oral Presentation

Workshop on Computer Vision and Generative Models for Medical Imaging, ICIAP 2025, Rome, Italy

Academic service

Reviewer

Journals

  • IEEE Transactions on Multimedia
  • IEEE Transactions on Industrial Informatics
  • Computer Vision and Image Understanding
  • Neural Computing and Applications
  • Applied Soft Computing

Conferences

  • International Joint Conference on Artificial Intelligence (IJCAI–ECAI)
  • British Machine Vision Conference (BMVC)
  • International Conference on Computer Vision Theory and Applications (VISAPP)

Student co-supervision

MSc thesis
2025 - Present

Fateh Ali

2025 - 2026

Navid Bakzadegan Tabrizi

2025 - 2026

Lokesh Kumar

2024 - 2025

Shakiba Sharifi

2023 - 2024

Farrukh Ullah