MVIP2026 - News: Keynotes, Panels and Workshops Announced

Keynotes, Panels and Workshops Announced

The organizing committee of the 14th National and 4th International Conference on Machine Vision and Image Processing (MVIP 2026) is pleased to announce the preliminary program. The conference will be held at the University of Kashan, Iran, on Wednesday–Thursday, November 11–12, 2026 (20–21 Aban 1405). All times are Iran Standard Time.See the keynote speakers' talks and biographies →MVIP 2026 keynote speakersKeynote SpeakersWednesday, Nov 11, 10:00–11:00 — Dr. Seyed Jalaleddin Mousavirad (Senior Member, IEEE; Assistant Professor, Mid Sweden University, Sweden): “When Algorithms Evolve: The Evolutionary Path to Better Vision Models.” The talk examines how evolutionary computation can automate neural architecture design, hyperparameter tuning and preprocessing optimization.Wednesday, Nov 11, 15:30–16:30 — Dr. Maryam Haji Esmaeili (Head of the Innovation Center, Refah Bank; faculty member, Islamic Azad University, Central Tehran Branch; active in fintech, AI and machine vision; image processing graduate of Kingston University London, with postdoctoral work at George Washington University).Thursday, Nov 12, 10:00–11:00 — Dr. Bhavesh Kataria (Associate Professor, School of Computing and Technology, Institute of Advanced Research, Gandhinagar, India): “From Pixels to Knowledge: Vision-Language Models for Digital Pathology and Biomedical AI.” The keynote explores how foundation models and VLMs can link whole-slide images with pathology reports, scientific literature and clinical knowledge.Thursday, Nov 12, 15:30–16:30 — Dr. Hamid Abrishami Moghaddam (Professor of Biomedical Engineering, K.N. Toosi University of Technology): “Artificial Intelligence in Lung CT Image Analysis: Advances and Priorities.”Specialized PanelsPanel 1: Machine Vision, Entrepreneurship, Governance, Ethics and Intelligent Management (Wed, 11:30–13:00)Panel 2: Machine Vision in Industry (Wed, 14:00–15:30)Panel 3: Machine Vision in Transportation, Traffic and Safe City Management (Wed, 16:30–18:00)Panel 4: Vision-Language Models (VLM) (Thu, 11:30–13:00)Panel 5: Machine Vision in Medicine and Digital Health (Thu, 14:00–15:30)Other Program HighlightsOral sessions (Hall A) and poster sessions (Hall B) covering image and video restoration and compression, industrial inspection and remote sensing, intelligent transportation, vision-language models, medical imaging, face recognition and biometrics, and human activity analysisA dedicated session for international students (Wed, 14:00–15:30)Industrial achievements and a short Product Talk (Thu, 08:30–10:00)Opening of the specialized exhibition (Wed, 11:00) and a ceremony honoring pioneers and distinguished figures of machine vision and image processing in Iran (Thu, 16:30)Best paper awards, competition announcements and the closing ceremony (Thu, 17:00–18:00)MVIP 2026 program
Click the program to open it full sizeWorkshopsFour workshops will accompany the scientific program: Vision-Language Models (VLMs) in Practice; Machine Vision and Autonomous Vehicles; Machine Vision in the Steel Industry; and Machine Vision in Medicine.The four-hour workshop “Vision-Language Models in Practice: From CLIP and SigLIP to Fine-Tuning with Unsloth” will be led by Alireza Akhavanpour, university lecturer, AI consultant and former head of the AI team at the knowledge-based company Shenasa. Topics include:VLM architectures, embedding-based versus generative models, and industrial applications such as image search, visual question answering, OCR and document analysis, and quality controlContrastive learning in CLIP and SigLIP (softmax vs. sigmoid loss) and their use in zero-shot classification and image-text retrievalRunning open-source VLMs, extracting embeddings and building a semantic image search systemFine-tuning with Unsloth: dataset preparation, LoRA and QLoRA, memory-efficient training on limited GPUs, evaluation and model exportThe workshop targets machine learning and computer vision enthusiasts with basic Python and PyTorch knowledge. Akhavanpour has over nine years of experience in AI projects across financial services, telecom and healthcare, has taught deep learning, LLMs and VLMs since 2016, and has delivered specialized courses for organizations such as MCI (Hamrah-e Avval) and the Central Bank of Iran. His courses have reached more than 100,000 students.Workshop dates and venues will be announced soon. To register or ask questions, contact the secretariat at mvip2026.kashan@gmail.com.Websites: mvip2026.ismvip.ir and mvip2026.kashanu.ac.ir
Note: this program is preliminary and subject to change.


The organizing committee of the 14th National and 4th International Conference on Machine Vision and Image Processing (MVIP 2026) is pleased to announce the preliminary program. The conference will be held at the University of Kashan, Iran, on Wednesday–Thursday, November 11–12, 2026 (20–21 Aban 1405). All times are Iran Standard Time.See the keynote speakers' talks and biographies →MVIP 2026 keynote speakersKeynote SpeakersWednesday, Nov 11, 10:00–11:00 — Dr. Seyed Jalaleddin Mousavirad (Senior Member, IEEE; Assistant Professor, Mid Sweden University, Sweden): “When Algorithms Evolve: The Evolutionary Path to Better Vision Models.” The talk examines how evolutionary computation can automate neural architecture design, hyperparameter tuning and preprocessing optimization.Wednesday, Nov 11, 15:30–16:30 — Dr. Maryam Haji Esmaeili (Head of the Innovation Center, Refah Bank; faculty member, Islamic Azad University, Central Tehran Branch; active in fintech, AI and machine vision; image processing graduate of Kingston University London, with postdoctoral work at George Washington University).Thursday, Nov 12, 10:00–11:00 — Dr. Bhavesh Kataria (Associate Professor, School of Computing and Technology, Institute of Advanced Research, Gandhinagar, India): “From Pixels to Knowledge: Vision-Language Models for Digital Pathology and Biomedical AI.” The keynote explores how foundation models and VLMs can link whole-slide images with pathology reports, scientific literature and clinical knowledge.Thursday, Nov 12, 15:30–16:30 — Dr. Hamid Abrishami Moghaddam (Professor of Biomedical Engineering, K.N. Toosi University of Technology): “Artificial Intelligence in Lung CT Image Analysis: Advances and Priorities.”Specialized PanelsPanel 1: Machine Vision, Entrepreneurship, Governance, Ethics and Intelligent Management (Wed, 11:30–13:00)Panel 2: Machine Vision in Industry (Wed, 14:00–15:30)Panel 3: Machine Vision in Transportation, Traffic and Safe City Management (Wed, 16:30–18:00)Panel 4: Vision-Language Models (VLM) (Thu, 11:30–13:00)Panel 5: Machine Vision in Medicine and Digital Health (Thu, 14:00–15:30)Other Program HighlightsOral sessions (Hall A) and poster sessions (Hall B) covering image and video restoration and compression, industrial inspection and remote sensing, intelligent transportation, vision-language models, medical imaging, face recognition and biometrics, and human activity analysisA dedicated session for international students (Wed, 14:00–15:30)Industrial achievements and a short Product Talk (Thu, 08:30–10:00)Opening of the specialized exhibition (Wed, 11:00) and a ceremony honoring pioneers and distinguished figures of machine vision and image processing in Iran (Thu, 16:30)Best paper awards, competition announcements and the closing ceremony (Thu, 17:00–18:00)MVIP 2026 program
Click the program to open it full sizeWorkshopsFour workshops will accompany the scientific program: Vision-Language Models (VLMs) in Practice; Machine Vision and Autonomous Vehicles; Machine Vision in the Steel Industry; and Machine Vision in Medicine.The four-hour workshop “Vision-Language Models in Practice: From CLIP and SigLIP to Fine-Tuning with Unsloth” will be led by Alireza Akhavanpour, university lecturer, AI consultant and former head of the AI team at the knowledge-based company Shenasa. Topics include:VLM architectures, embedding-based versus generative models, and industrial applications such as image search, visual question answering, OCR and document analysis, and quality controlContrastive learning in CLIP and SigLIP (softmax vs. sigmoid loss) and their use in zero-shot classification and image-text retrievalRunning open-source VLMs, extracting embeddings and building a semantic image search systemFine-tuning with Unsloth: dataset preparation, LoRA and QLoRA, memory-efficient training on limited GPUs, evaluation and model exportThe workshop targets machine learning and computer vision enthusiasts with basic Python and PyTorch knowledge. Akhavanpour has over nine years of experience in AI projects across financial services, telecom and healthcare, has taught deep learning, LLMs and VLMs since 2016, and has delivered specialized courses for organizations such as MCI (Hamrah-e Avval) and the Central Bank of Iran. His courses have reached more than 100,000 students.Workshop dates and venues will be announced soon. To register or ask questions, contact the secretariat at mvip2026.kashan@gmail.com.Websites: mvip2026.ismvip.ir and mvip2026.kashanu.ac.ir
Note: this program is preliminary and subject to change.
Highlights
Return to News Page