July 16th, 2024 
Daniela Lopez de Luise Towards Multimodal Foundational Models for Modelling Visual Cortex
Held in conjunction with the 18th European Conference on Computer Vision ECCV 2024 – Sep 29th – Oct 4th, 2024, Milan (Italy)
IMPORTANT DATES
Paper submission deadline: July 15, 2024 extended to July 22, 2024
Notification to authors: August 10, 2024
Camera-ready deadline: August 20, 2024
CALL FOR PAPERS
Recent advancements in artificial intelligence and computational modeling have sparked a new era in understanding complex systems, particularly in neuroscience. Foundation models, distinguished by their capability to integrate diverse data modalities, have emerged as powerful tools in decoding the complexities of the visual cortex in both human and animal subjects. This workshop aims to explore state-of-the-art techniques and methodologies for developing multimodal foundational models that comprehensively represent the visual cortex.
Topics (but not limited to):
-
Theoretical Frameworks and Computational Approaches: Novel theoretical constructs and computational strategies for modeling the visual cortex using multimodal data.
-
Integration of Diverse Data Sources: Techniques and challenges in integrating and harmonizing heterogeneous data modalities such as fMRI, EEG, in vivo two-photon calcium imaging, fNIRS, and others.
-
Learning Paradigms for Noisy Data: Innovations in learning algorithms and paradigms to effectively handle noisy and incomplete data in modeling brain functions.
-
Applications in Neuroscientific Research: Practical applications of multimodal foundational models in elucidating perception, cognition, and neurodevelopmental or neurodegenerative disorders.
-
Contrastive Learning for Multimodal Brain Data Fusion: Techniques and advancements in leveraging contrastive learning methods to fuse multimodal brain data for enhanced representation and analysis.
-
Self-Supervised Learning for Temporal Brain Dynamics: Approaches utilizing self-supervised learning to capture and model temporal dynamics in brain imaging and physiological data.
-
Unsupervised Learning for Structural and Functional Brain Network Construction: Methods employing unsupervised learning to construct and analyze structural and functional brain networks from multimodal data.
-
Weakly Supervised Learning for Brain Connectivity Analysis: Innovations in weakly supervised learning techniques for analyzing brain connectivity patterns and networks.
-
Foundational Models for Classification and Predictive Modeling: Development and application of foundational models for classification and predictive modeling tasks in neuroscience.
-
Multimodal Brain Image Visualization with Advanced Learning Techniques: Techniques for visualizing multimodal brain images using advanced learning and visualization methods to aid in data interpretation.
-
Ethical Implications: Ethical considerations in the creation, use, and implications of foundational models in neuroscience research and applications.
SUBMISSION GUIDELINES:
We invite original contributions in the form of full research papers, review articles, or extended abstracts of ongoing works aligned with the workshop's themes. Full papers should not exceed 8 pages (excluding references), while extended abstracts should be limited to 4 pages. All submissions will undergo a rigorous peer-review process conducted by experts in the field.
Submission follows the ECCV format (see the ECCV website for format instruction (https://eccv.ecva.net/Conferences/2024/AuthorGuide)).
WORKSHOP PROCEEDINGS
All accepted contributed papers will appear in the ECCV2024 workshop proceedings.
WORKSHOP ORGANIZERS
Nouria Lakhdar Ghazal – Mohammed V University of Rabat, Morocco
Simone Palazzo – University of Catania, Italy
Matteo Pennisi – Campus Bio Medico Rome, Italy
Federica Proietto Salanitri – University of Catania, Italy
Shanmuganathan Raman – Indian Institute of Technology, India
Concetto Spampinato – University of Catania, Italy
Andreas Tolias – University of Houston, Texas
Jonathan Xu – University of Waterloo, Canada
Posted in