Photo

Simon Thomine

I am a Ph.D. student in computer vision and deep learning at the University of Technology of Troyes. My current research focuses on industrial unsupervised anomaly detection, particularly in textures. In addition to my academic pursuits, I am actively working towards making AI accessible to a wider audience through personal projects.

🤗

My projects 🛠

πŸš€ Apprendre le Deep Learning Γ  partir de zΓ©ro πŸš€

This project is a French-language course designed to teach deep learning from scratch, requiring only basic Python and mathematical knowledge. It aims to provide a comprehensive introduction to the core principles of deep learning, all within a single GitHub repository regrouping various notebook-based lessons. I believe that making AI accessible to a wider audience is crucial in today's world, given the growing presence of AI in our daily lives.

course banner

🏭 Industrial Defect Generator 🏭

This project introduce a new library dessigned to generate synthetic defects based on proposed methods from the litterature.

Nsa defect generation example
🤗

🧡 Industrial textile dataset 🧡

This project introduces a new fabric anomaly dataset with 10 textile categories captured using high-resolution factory cameras.

Examples of defect-free and defective images from ITD

πŸ­βš— Distillation Industrial Anomaly Detection βš—πŸ­

This project regroups advanced methods utilizing knowledge distillation for unsupervised anomaly detection.

STPM figure

Implementation of Remembering Normality

Implementation of the paper "Remembering Normality: Memory-guided Knowledge Distillation for Unsupervised Anomaly Detection"

RN figure
📜

Publications 📚

CSE: Surface Anomaly Detection with Contrastively Selected Embedding 📜

Thomine, Simon and Hichem Snoussi. "CSE: Surface Anomaly Detection with Contrastively Selected Embedding." 2024 International Conference on Computer Vision Theory and Applications (VISAPP 2024). 2024.

Distillation-based fabric anomaly detection 📜

Thomine, Simon, and Hichem Snoussi. "Distillation-based fabric anomaly detection." Textile Research Journal 94.5-6 (2024): 552-565.

Dual Model Knowledge Distillation for Industrial Anomaly Detection 📄

Thomine, Simon and Snoussi, Hichem, "Dual Model Knowledge Distillation for Industrial Anomaly Detection."" Pattern Analysis and Applications (accepted preprint)

FABLE: Fabric Anomaly Detection Automation Process 📜

Thomine, Simon, Hichem Snoussi, and Mahmoud Soua. "FABLE: Fabric Anomaly Detection Automation Process." 2023 International Conference on Control, Automation and Diagnosis (ICCAD). IEEE, 2023.

MixedTeacher: Knowledge Distillation for fast inference textural anomaly detection 📜

Thomine, Simon, Hichem Snoussi, and Mahmoud Soua. "MixedTeacher: Knowledge Distillation for fast inference textural anomaly detection." 2023 International Conference on Computer Vision Theory and Applications (VISAPP 2023). 2023.

Skills πŸ’‘

πŸ“š Theoritical πŸ“š

Convolutional Neural Networks, Knowledge Distillation, Vision Transformer, Object Tracking, LLM/VLM, Application Development, Classical Computer Vision

πŸ’» Software πŸ’»

Python, C++, Pytorch, Transformers, Torchvision, OpenCV, Numpy, Pillow, Timm, Scikit-learn, Scikit-image, QT, TensorRT, Git, Docker

πŸŽ“ Education πŸŽ“

Several teaching experiences at University of Technology of Troyes and ESTP.

πŸ—£οΈ Languages πŸ—£οΈ

French (fluent), English (proficient), Spanish (basic)

Education 🎓

πŸŽ“ Ph.D in Computer Vision and Artificial Intelligence

utt 🏫 University of Technology of Troyes
πŸ“… Planned to September 2024
πŸ“ Visual Anomaly Detection on Surfaces through Unsupervised Learning for the Future Industry

πŸ”§ Engineer's degree in Application Development for Image Processing

tse laval
🏫 Télécom Saint-Etienne and Université Label
πŸ“… September 2020
πŸ“ Image application development

tse

tse

Miscellaneous 💬

French "Three Minute Thesis" presentation

mt180
tse

Contact 📞


🤗