Laboratory of Imaging, Vision and Artificial Intelligence (LIVIA)

The Laboratory of Imaging, Vision and Artificial Intelligence (LIVIA) is a research unit at the École de technologie supérieure (ÉTS) in Montreal, focusing on the application of artificial intelligence (AI) to solve complex real-world problems. LIVIA's work centres around the visual perception of 2D and 3D environments, using AI for dynamic and static modelling. Their research areas include machine learning, computer vision, pattern recognition, adaptive and intelligent systems, information fusion, and optimisation of complex systems.

LIVIA's primary applications cover medical and satellite imaging, video analysis, surveillance, biometrics (such as face and voice recognition), affective computing in healthcare, and document digitisation. A key focus is on developing AI-based systems capable of handling large datasets with limited annotations, for tasks like tumour detection or assessing mental health conditions.

LIVIA also conducts significant research and development, leading to innovations like convolutional neural networks and advanced AI algorithms that have practical applications in fields such as healthcare, security, and robotics. The lab has a strong reputation for its contributions to AI engineering, both academically and industrially, as seen in numerous publications and collaborations.

LIVIA is associated with several research chairs, including those on embedded neural networks for building control, computer vision for industrial applications, and AI for digital health. The laboratory has been operational for nearly 30 years and is recognised for training highly qualified personnel and disseminating research through international conferences and journals. It also ranks 6th in Canada for computer vision research, according to CSRankings.

The Laboratory of Imaging, Vision, and Artificial Intelligence (LIVIA) at École de Technologie Supérieure (ÉTS) in Montreal was established in the early 1990s. The lab was founded to focus on applying artificial intelligence (AI) to solve complex problems related to image processing, computer vision, and environmental modelling. From its inception, LIVIA aimed to address practical challenges by developing AI solutions for 2D and 3D visual perception and static and dynamic scene modelling.

In the 1990s, LIVIA’s early work revolved around foundational AI research, including machine learning and computer vision. During this period, the lab focused on AI-based techniques for large-scale image and video processing. Early projects included the development of methods for satellite and medical image analysis. The lab also worked on biometric systems, using AI to recognise individuals through facial features, voice, and signatures.

In the 2000s, LIVIA expanded its research to incorporate more advanced machine learning techniques, including pattern recognition and adaptive intelligent systems. A major breakthrough during this time was the lab’s development of convolutional neural networks (CNNs), a significant innovation in AI that allowed for the detection of objects like tumours in medical images. This marked the beginning of LIVIA’s involvement in the healthcare sector, where AI was applied to diagnose medical conditions using image data.

During this decade, LIVIA also contributed to research in affective computing, focusing on the assessment of human emotions through non-verbal cues, such as facial expressions and voice, for healthcare applications. This led to advancements in personalised health monitoring systems, which aimed to track patients’ emotional and mental health in real time.

By the 2010s, LIVIA had established itself as a leading research lab in AI. It partnered with industry and academia to advance its work in AI-driven solutions. LIVIA became associated with key research chairs, including the Distech Controls Industrial Chair on Embedded Neural Networks for Connected Building Control and the Matrox Imaging Industrial Research Chair in Computer Vision for Industrial Applications. These partnerships allowed the lab to extend its expertise into industrial applications, developing AI algorithms for tasks such as object detection, optical character recognition, barcode reading, and intelligent building management systems.

In 2015, LIVIA’s research reached new heights with the application of AI in neurotechnology, including brain-machine interfaces. This work involved using AI to interpret signals from the human brain and developing systems that could interface directly with neural activity, opening up new possibilities in healthcare and cognitive research.

LIVIA’s educational mission also gained traction during these years. The lab contributed to the training of a large number of graduate and post-graduate students, as well as postdoctoral fellows. LIVIA’s researchers published extensively in internationally recognised journals and conferences, with hundreds of publications advancing the state of AI research globally.

In 2019, LIVIA’s work expanded further into security and surveillance systems, with AI playing a central role in facial recognition and video analysis technologies. This added a new dimension to the lab’s applications in safety and public security, making LIVIA a key player in the development of AI-based surveillance systems.

Throughout its history, LIVIA has been actively involved in numerous research and development projects. One of its primary strengths has been the ability to process massive quantities of data with limited annotations, especially in medical imaging and satellite data analysis. The lab’s work has also extended to the automatic processing of handwritten documents, helping to modernise document management systems through AI.

By 2023, ÉTS and LIVIA’s contributions to AI research had earned the university a strong position in global rankings. LIVIA was instrumental in ÉTS being ranked 6th in Canada for computer vision research, as listed by CSRankings. The lab continues to conduct cutting-edge research in AI, with ongoing projects focusing on healthcare applications, neurotechnology, document processing, and security systems.

Today, LIVIA remains a leader in AI research, with strong partnerships in both academic and industrial sectors. Its contributions to AI development continue to solve real-world problems in healthcare, surveillance, and industrial automation, ensuring that LIVIA plays a vital role in the future of AI research and innovation.

The Laboratory of Imaging, Vision, and Artificial Intelligence (LIVIA) at École de Technologie Supérieure (ÉTS) offers a range of specialised courses focusing on artificial intelligence (AI), computer vision, machine learning, and related fields. These courses are designed to provide students with both theoretical knowledge and practical skills in AI technologies, preparing them for careers in research and industry.

• Machine Learning: This course introduces students to the fundamental concepts of machine learning, which involves designing systems that can learn from data. It covers supervised and unsupervised learning, reinforcement learning, and neural networks. Students learn how to build machine learning models, evaluate their performance, and apply them to real-world problems. The course also includes practical projects where students use machine learning algorithms in areas like healthcare and image recognition.
• Computer Vision: In this course, students explore the field of computer vision, which enables machines to interpret and understand visual information from the world. The course covers key topics such as image processing, feature extraction, object detection, and 3D scene reconstruction. Students work with tools and frameworks used in computer vision, gaining hands-on experience in analysing images and videos. Applications include medical imaging, satellite data analysis, and surveillance systems.
• Pattern Recognition: Pattern recognition is a core AI field that focuses on the identification and classification of patterns in data. This course teaches students the techniques used in recognising patterns, such as statistical methods, machine learning algorithms, and deep learning. Students apply these techniques to problems in biometrics, such as face and voice recognition, and handwritten document processing.
• Affective Computing: This course focuses on AI systems that can detect and respond to human emotions. It covers methods for recognising non-verbal cues like facial expressions and voice intonations. Students learn to design AI systems that monitor mental health and emotional well-being, which is increasingly used in personalised healthcare applications.
• Information Fusion: Information fusion involves combining data from multiple sources to make better decisions. This course teaches students how to merge data from sensors, cameras, and other inputs to improve the accuracy and reliability of AI systems. Applications include autonomous vehicles, robotics, and security systems. Students work on practical projects that involve data fusion for real-time decision-making.
• Adaptive and Intelligent Systems: This course covers systems that can adapt to their environments and improve their performance over time. Students learn about AI algorithms that allow machines to adjust to new information and changing conditions. Applications include robotics, where machines need to adapt to dynamic environments, and industrial automation, where AI is used to optimise processes.
• Deep Learning: A specialised course in deep learning teaches students how to design and train neural networks, with a focus on convolutional neural networks (CNNs) and recurrent neural networks (RNNs). The course covers the application of deep learning in fields such as medical image analysis, natural language processing, and automated diagnosis systems. Students engage in practical projects, working with large datasets to develop AI models for complex tasks.
• Optimisation of Complex Systems: This course focuses on optimising AI systems to improve efficiency and accuracy. Students learn how to apply optimisation techniques to complex problems such as supply chain management, resource allocation, and industrial systems. The course also explores evolutionary algorithms and how they can be used to find optimal solutions in real-world applications.

• ÉTS is ranked 6th in Canada for computer vision research according to CSRankings.
• LIVIA’s research and contributions in artificial intelligence and machine learning have placed ÉTS among the top institutions globally in these fields.
• ÉTS is recognised for its focus on applied engineering and technological research, giving it a strong reputation both nationally and internationally.
• LIVIA's partnerships with industry and academic institutions contribute to ÉTS’s standing in global rankings related to AI research and technology development.

École de Technologie Supérieure (ÉTS) has an impressive job integration rate for its graduates, including those from the Laboratory of Imaging, Vision, and Artificial Intelligence (LIVIA). On average, 92% of graduates secure employment within six months of graduation. Many students benefit from the university’s strong industry connections, leading to internships and full-time placements in sectors like healthcare, AI research, industrial automation, and security. Over 1,000 placements are facilitated annually through partnerships with companies and research projects, ensuring students have excellent opportunities to enter the workforce in their chosen fields.

• LIVIA - Imaging, Vision and Artificial Intelligence Laboratory| ÉTS Montréal
• Research - Montréal| ÉTS Montréal
• Livia | AI Directory - Global Artificial Intelligence| IT World Canada
• Top Artificial Intelligence Research Labs in the World| LinkedIn · Amer Kareem
• Artificial Intelligence Research Labs - AI Engineer Guide| aiengineer.guide
• Top 10 AI Research Labs Worldwide| Datatechvibe
• Enstitüler| Türkiye Yapay Zeka ?nisiyatifi
• RESEARCH AND INNOVATION IN AI AT ÉTS Knowledge| Innovations of the World
• TFS-ViT: Token-level feature stylization for domain| ScienceDirect.com
• Distech Controls Collaborates with École de Technologie| Acuity Brands Insights
• Jose Dolz's Homepage| Jose Dolz
• Fully-Funded 4 years PhD Position at ETS Montreal| UAB Barcelona