Our invited speakers come from top research institutions and companies around the globe, and are leading figures in the topics covered by the workshop. This diverse selection will prove valuable for academic as well as industry researchers and practitioners. Both practical and theoretical aspects of quantum computing and its use in computer vision will be covered by the invited lecturers. We will make a selection out of the confirmed speakers, with a potential to include unconfirmed speakers. Both of these are listed below.
Key Information:
What is Quantum Computing?
Quantum computing (QC) is an emerging computing paradigm that harnesses quantum-mechanical phenomena, such as superposition and entanglement, to process information. Unlike classical computers that operate on binary bits, quantum computers use qubits, enabling fundamentally different approaches to optimization, simulation, and sampling. While today's quantum devices remain in the Noisy Intermediate-Scale Quantum (NISQ) era, rapid advances in both hardware and algorithms are making practical QC increasingly accessible.
What is Quantum Computer Vision & Machine Learning?
Quantum Computer Vision and Machine Learning (QCVML) explores how QC can complement modern AI systems. Rather than replacing deep learning, quantum algorithms are designed to work alongside classical models, enabling hybrid quantum–classical workflows. Current research investigates applications including optimization, geometric estimation, clustering, graph matching, image generation, 3D scene representation, and multimodal learning.
Why Quantum Generative Models?
Generative AI has become a cornerstone of modern computer vision, powering diffusion models, image synthesis, video generation, neural scene representations, and large multimodal models. At the same time, these models require enormous computational resources for training and inference. Quantum generative models—including quantum GANs, quantum diffusion models, Born machines, and hybrid quantum–classical architectures—are emerging as a promising research direction for learning and sampling high-dimensional distributions, with the potential to improve computational efficiency and scalability as quantum hardware matures.
Why Now?
Computer vision has repeatedly been transformed by new computing hardware—from CPUs to GPUs and AI accelerators. QC represents the next frontier. Significant progress in quantum hardware, software frameworks, and algorithms has enabled the first demonstrations of practical computer vision applications on both quantum annealers and gate-based quantum processors. Meanwhile, the computational and energy demands of large-scale generative AI continue to grow rapidly, making this an ideal time to explore where QC can provide genuine advantages and establish realistic benchmarks for future research.
Workshop Vision?
The 3rd Workshop on Quantum Computer Vision and Machine Learning (QCVML) provides an accessible entry point for the ECCV community into this rapidly growing field. The workshop spans both quantum annealing and gate-based QC, with a special focus on quantum generative models for vision and multimodal learning. Through invited talks, tutorials, and discussions, QCVML aims to connect researchers from computer vision, machine learning, QC, physics, and industry, fostering collaboration and shaping the future of quantum-enhanced vision systems.
Why Attend?
We have a packed and exciting day ahead of us!
We gladly welcome you to our workshop, QCVML 2026.
Quantum-enhanced Computer Vision: Going Beyond Classical Algorithms
Quantum computing has attracted growing interest in computer vision, promising new computational paradigms for optimization, learning, and probabilistic inference, and giving rise to the emerging field of quantum-enhanced computer vision (QeCV). This talk provides an accessible overview of the current landscape of QeCV, covering its foundations, recent advances, emerging research directions, and limitations.
We begin by motivating the use of quantum computing for computer vision and reviewing recent algorithmic advances, as well as realistic hardware and development timelines. We then introduce the fundamental principles of quantum computing, contrasting gate-based and adiabatic paradigms, and discuss their relevance for vision applications, along with practical software tools and programming frameworks available to the community. Finally, we survey emerging application areas, including combinatorial optimization, continuous neural representations, and probabilistic generative models based on quantum sampling. We conclude with a brief discussion of current challenges, including hardware limitations, data encoding, and trainability. Overall, the talk provides researchers with a balanced perspective on the current state of QeCV and its future research opportunities.
Optimization on Ion-Trap Hardware
eleQtron is a quantum computing startup specializing in ion-trap technology. To address individual ions, it employs a unique approach known as MAGIC (Magnetic Gradient Induced Coupling). This technique enables precise control of the quantum states of ions using established and highly miniaturized microwave technology. A big advantage of this approach is also that there is all-to-all connectivity between the qubits in a register. In this talk the possibility of implementing quantum annealing protocols on hardware like this is discussed. The optimization problem we investigate stems from multi object tracking.
We invite you to join us for a brief overview of recently published papers on quantum computer vision.
Towards more Pervasive Quantum Computing: from Quantum Error Mitigation to Graph Latent Spaces
This talk will focus on his recent research towards making near-term quantum computing immediately more pervasive and practical. The talk will focus mainly on two fronts: 1) Quantum Error Mitigation (QEM) using a newly proposed attention-based graph neural network and multi-layer perceptron (QTransMLP) [1,2,3,5,6]. Our approach encodes quantum circuits using a graph transformer, with a multi-head attention that captures global features of the circuits as well as local lightcone mask features and estimate noise-mitigated outputs. It removes the overhead burden of classical QEM methods, such as Zero Noise Extrapolation (ZNE), while outperforming the state-of-the-art methods; and 2) development of Graph Variational Latent Spaces (GVLS) to compress large graphs into representations compact enough for a Quantum Graph Neural Network (QGNN) [4,7,8]. Such latent space can encode large input graphs into smaller latent ones so that qubit and q-gate requirements scale with the compressed representation.
[1] Tousi, S. M. A., Bazgir, A., Zhang, Y., & DeSouza, G. N. (2026). Solving Burgers Equation on Noisy Quantum Hardware with Error Mitigation through Attention Graph Transformers. AIP Physics of Fluids Journal. (under revision)
[2] Tousi, S. M. A., Bazgir, A., Zhang, Y., & DeSouza, G. N. (2025). Quantum Error Mitigation with Attention Graph Transformers for Burgers Equation Solvers on NISQ Hardware. arXiv preprint arXiv:2512.23817. (pre-print)
[3] Tousi, S. M. A., & DeSouza, G. N. (2025). QAGT-MLP: An Attention-Based Graph Transformer for Small and Large-Scale Quantum Error Mitigation. arXiv preprint arXiv:2511.03119 and submitted to Springer Nature Communications AI & Computing. (under revision)
[4] Tousi, S. M. A., & DeSouza, G. N. (2026). Neural Networks for Quantum Computing: A Study of How Neural Networks Can Address Quantum Limitations. arXiv preprint.
[5] Tousi, S. M. A., & DeSouza, G. N. (2026). Comparing Simulated versus Hardware Training Data for Learning-Based Quantum Error Mitigation. In the Proceedings of IEEE International Joint Conference on Neural Networks 2026. June 2026.
[6] Tousi, S. M. A., Bazgir, A., Zhang, Y., & DeSouza, G. N. (2025). Physics-Informed Graph Neural Network Error Correction for Quantum CFD and PDE Solvers on NISQ Hardware. AIP Physics of Fluids Journal. (under review)
[7] Sara Shojaei, Seyed Mohamad Ali Tousi, Emma Bennett, Param Sangani, Ali Shiri Sichani, Ilker Ersoy, Hadi Ali-Akbarpour, Filiz Bunyak, and G. N. DeSouza, "Quantum-Enhanced Similarity Measures for Polarimetric Materials Classification". arXiv preprint.
[8] Tousi, S. M. A., & DeSouza, G. N. (2026). Graph Variational Latent Space (GVLS) for Quantum Graph Neural Networks, IEEE Transactions on Artificial Intelligence. (in preparation)
Untill next time!
QCVML 2026 does not host a paper or poster submission track. Instead, we continue the successful Highlight Talks session from previous editions, providing a platform to showcase and discuss recent advances in quantum generative models for computer vision and machine learning.
Authors of recently published, peer-reviewed papers are encouraged to contact the organizing committee to propose a highlight talk. Topics of interest include, but are not limited to:
Preference will be given to recent publications at leading venues in computer vision, machine learning, quantum computing, and quantum information. Interested authors are encouraged to contact the organizing committee at frances.yfy@gmail.com with a brief summary of their work and publication details.
Here are the diligent people behind QCVML 2026.
A small community in Quantum Computer Vision has formed over the last years and is steadily growing. With quantum computers being easily accessible by now, we believe that the time is right to show the potential of quantum computer vision and to bring together researchers from both fields.
Attending the workshop is a great opportunity to gain a glimpse on the basics in quantum computing. Furthermore, invited talks from academia and industry will give an overview of the current developments of the field and our poster session will give you the opportunity to see the current state-of-the-art collected in one place.
Yes, QCVML will be held fully in-person in September at ECCV 2026. While there will be no live streaming, we will record the talks and make them available on the website. Furthermore, an extended version of the poster session will be presented on the website.
Follow the recent hapennings here.
