in conjunction with ISCAS’2025, London 25-28 May 25–28 2025
MUSCLE (formerly IVU, Image and Vision Understanding) is the ERCIM Working Group focused on multimedia understanding through semantics, computation, and learning. For over 15 years, this working group has brought together teams from ERCIM and non-ERCIM institutions, uniting expertise in machine learning, artificial intelligence, image/video/audio processing, and multimedia processing and management.
The International Workshop on Computational Intelligence for Multimedia Understanding (IWCIM) is the group’s flagship annual event. Following the 2024 edition, held in conjunction with ISCAS in Singapore, and the 2023 edition, co-located with ICASSP in Rhodes, Greece, the 2025 workshop will again be hosted in collaboration with ISCAS. This time, it will take place in London from May 25–28, 2025.
The primary aim of IWCIM 2025 is to present and discuss current trends and future directions in computational intelligence for multimedia understanding, while fostering the creation of a robust network of scientists and practitioners. This network aims to provide seamless access to expertise, data, and ideas.
Multimedia understanding is a cornerstone of many intelligent applications that influence our daily lives, from household devices to commercial, industrial, service, and scientific domains. Analyzing raw data to extract semantics is vital for unlocking their full potential and enabling us to manage everyday tasks efficiently. Today’s raw data originate from a wide variety of sensors and sources, differing in nature, format, reliability, and information content. Multimodal and cross-modal analysis is indispensable to harnessing these data effectively. Beyond data analysis, these challenges extend to data description, supporting efficient storage and mining. The interoperability and exchangeability of heterogeneous and distributed data are crucial for practical applications. Semantics represents the highest level of information. Inferring semantics from raw data requires leveraging both the data itself and prior knowledge to extract structure and meaning. Computational methods, including machine learning, statistical modeling, and Bayesian approaches, are essential to achieve this goal at various levels.
The scope of IWCIM 2025 encompasses, but is not limited to, the following topics:
• Multisensor systems
• Multimodal analysis
• Crossmodal data analysis and clustering
• Mixed-reality applications
• Activity and object detection and recognition
• Text and speech recognition
• Multimedia labeling, semantic annotation and metadata
• Multimodal indexing and searching in very large data-bases
• Big and Linked Data
• Search and mining Big Data
• Large-scale recommendation systems
• Multimedia and Multi-structured data
• Cloud Optimization
• Pervasive intelligence
• Machine learning in multimedia understanding
• Attention based approaches for multimedia understanding
• Diffusion models for multi-modal data analysis
• Multi-modal data analysis in compressed domain
• Multi-modal data analysis for remote sensing applications
• Semantic web and Linked data
• Case studies.
Link: http://iwcim.itu.edu.tr
Please contact:
Behçet Uğur Töreyin, ITU, Istanbul,Turkey
Maria Trocan, Institut Supérieur d'Électronique de Paris (Isep), Paris, France
Davide Moroni, ISTI-CNR, Pisa, Italy
