|
|
|
PROGRAM
 |
Wed. 21 | Thu. 22 | Fri. 23 |
|
08:00
09:00
10:00
11:00
12:00
13:00
14:00
15:00
16:00
17:00
18:00
19:00
20:00
21:00
22:00
|
8:30 - 8:45 (15min)
Registration
Registration Desk - Main Hall
8:45 - 9:00 (15min)
Opening
 Amphitheater
Program Chairs
9:00 - 9:50 (50min)
Opening Keynote
Amphitheater
Marco BERTINI
9:50 - 10:15 (25min)
Poster and Demo
Main Hall
10:15 - 10:30 (15min)
Coffee break
Main Hall
10:30 - 11:30 (1h)
[MDFSA] Multimodal Data Analysis and Fusion for Smart Agriculture
Amphitheater
› Multisensor and Multitemporal Vegetation Stress Investigation with Earth Observation Missions and Field Data
- Giacomo Lazzeri, Agenzia Spaziale Italiana (ASI)
10:30-11:30 (1h)
› Edge-Aware Representation Learning for Efficient Crop Monitoring in Precision Agriculture
- Sarathchandrakumar Thottuchirayil Sasidharan, Sapienza - University of Rome
10:30-11:30 (1h)
› Multimodal Earth Observation Data Fusion for Wheat and Barley Agricultural Crop Discrimination
- Matei Debu, Transilvania University of Brasov
10:30-10:45 (15min)
› Explainability in Plant Disease Recognition: The Joint Role of Architecture and Domain Adaptation
- RADIA BAHAYA, National School of Artificial Intelligence and Digitalization, Mohammed First University of Oujda
10:30-11:30 (1h)
› Row by Row: A Transformer-based Approach for Vine Row Segmentation Across Diverse Vineyards
- Andreas Gallaun, Joanneum Research Forschungsgesellschaft mbH - Stefanie Onsori-Wechtitsch, Joanneum Research Forschungsgesellschaft mbH - Peter Schallauer, Joanneum Research Forschungsgesellschaft mbH
10:45-11:00 (15min)
› From Pixels to Agronomic Risk Narratives: Adapting Vision-Language Foundation Models to Agricultural Data
- Robert-Ionuț Vătășoiu, Polytechnic University of Bucharest [Romania] = Université Politehnica de Bucarest [Roumanie] = Universitatea Națională de Știință și Tehnologie Politehnica București [România], GEOSense, CAMPUS Institute - Teodor Costachioiu, Polytechnic University of Bucharest [Romania] = Université Politehnica de Bucarest [Roumanie] = Universitatea Națională de Știință și Tehnologie Politehnica București [România], GEOSense, CAMPUS Institute - Daniela Faur, Polytechnic University of Bucharest [Romania] = Université Politehnica de Bucarest [Roumanie] = Universitatea Națională de Știință și Tehnologie Politehnica București [România], GEOSense, CAMPUS Institute
11:00-11:15 (15min)
› Late Fusion of Vision and Text Descriptors for Improving Explainability of Crop Discrimination
- Andreea Nițu, National University of Science and Technology Politehnica of Bucharest - Corneliu Florea, Transilvania University of Brasov, National University of Science and Technology Politehnica of Bucharest - Mihai Ivanovici, Transilvania University of Brasov - Andrei Racoviteanu, National University of Science and Technology Politehnica of Bucharest
11:15-11:30 (15min)
11:30 - 12:30 (1h)
[MusiCHER] Challenges and opportunities on content-based multimedia indexing in the sphere of musical cultural heritage
Amphitheater
› Emotion-Driven Music Generation for Cultural Heritage Experiences
- Vasilis Papadopoulos, Catalink Limited - Georgia Christodoulou, Catalink Limited
11:30-11:45 (15min)
› Measurement-Based Acoustic Digitization of a Pipe Organ in Its Architectural Context
- Stephan Werner, Technische Universität Ilmenau = Ilmenau University of Technology = Université technique d'Ilmenau [Ilmenau, Allemagne]
11:45-12:00 (15min)
› Taste-aware music retrieval from audio embeddings
- Matteo Spanio, Università degli Studi di Padova = University of Padua
12:00-12:15 (15min)
› DIMA: Toward a Multimedia Inspection Workbench for Musical Cultural Heritage Records
- Andrea Gulli, Department of Musicology and Cultural Heritage, University of Pavia
12:15-12:23 (08min)
› From Digital Representations to Storyliving Toolkits for Virtual Worlds in Musical Cultural Heritage
- Eleftherios Anastasovitis, Centre for Research and Technology Hellas
12:23-12:30 (07min)
12:30 - 13:15 (45min)
Lunch
13:15 - 14:00 (45min)
Poster session
Main Hall
14:00 - 16:00 (2h)
Regular session 1
Amphitheater
› Beyond generic augmentations: contrastive learning for historical image clustering
- Raphaël Renard, LIP6
14:00-14:15 (15min)
› Guiding Zero-Shot Composed Image Retrieval through Semantic Understanding of Captions
- Pavan Kartheek Rachabathuni, Università degli Studi di Firenze = University of Florence = Université de Florence
14:15-14:30 (15min)
› Refinement of clinical process models based on the integration of physiological time series
- Hugo Boisaubert, IRIT
14:30-14:45 (15min)
› Self-Supervised Learning from FMCW Radar Signals for vehicles detection
- Yassine Janati, IMT Nord Europe
14:45-15:00 (15min)
› Physical Structure-Constrained Vision Language Model for Grounded Caption Enhancement
- Nguyen Nam, Nguyen Xuan Nam - Hidetomo Sakaino, Sakaino
15:00-15:15 (15min)
› RetroCLIP: Learning vision-language representations for historical photographs
- Enzo Bonneau, Laboratoire d'Informatique Paris Descartes
15:15-15:30 (15min)
› Hybrid Detection of Mountain Gorillas via Confidence-Gated Morphological Validation
- Kandolo Herman, International School, Vietnam National University (IS-VNU)
15:30-15:37 (07min)
› A Unified Framework for Gloss-Free Sign Language Translation with Contrastive Pretraining and Transformer Temporal Modeling
- Raditya Faisal Waliulu, Hiroshima University
15:37-15:45 (08min)
› Continuous Sign Language Augmentation from Isolated Signs using Landmark Blending
- Tiago Gonçalves, NOVA-LINCS ISEL, Instituto Superior de Engenharia de Lisboa
15:45-15:53 (08min)
16:00 - 16:30 (30min)
Coffee break
Main Hall
16:30 - 18:00 (1h30)
Demo Session + Music meets science
Diluvienne is an Occitan vocal trio that reinterprets the musical traditions of southern France with a modern, socially conscious, and inclusive approach, featuring warm polyphonic harmonies and original compositions.
18:00 - 19:30 (1h30)
Toulouse City visit
Let's discover the city
19:30 - 21:00 (1h30)
Welcome reception
Ma biche sur le toit
https://www.mabichesurletoit.com/
|
8:30 - 9:00 (30min)
Registration
Registration
9:00 - 9:30 (30min)
Women in AI
Amphitheater
Florence Sedes
9:30 - 10:30 (1h)
Regular session 2
Amphitheater
› A Cost-Aware Multimodal Routing Framework for Product Unification in Large-Scale E-Commerce Catalogs
- Uğur Berk Şahin, Trendyol Group, Data Science - Catalog Quality & Moderation - Harun Türkmen, Trendyol Group, Data Science - Catalog Quality & Moderation
09:30-09:45 (15min)
› A Data-Centric Multimodal Framework for Medical Visual Question Answering on VQA-RAD
- Mariem Abdedaiem, Isimg
09:45-10:00 (15min)
› A novel network for classification of cuneiform tablet metadata
- Hagelskjaer Frederik, University of Southern Denmark
10:00-10:15 (15min)
› A framework for French Medical Audio Transcription with Hybrid Error Correction
- Catherine PONS, ISIS Castres / INUC / IRIT - Yohann Chasseray, ISIS Castres / INUC / IRIT - Imen Megdiche, ISIS Castres / INUC / IRIT
10:15-10:23 (08min)
› CrossTrans-MFC: An Effective Cross Transformer Approach for Multimedia Fact-Checking
- Hoang Long Nguyen, Laboratoire d'Informatique Fondamentale et Appliquée de Tours
10:23-10:30 (07min)
10:30 - 10:45 (15min)
Coffee break
Main Hall
11:00 - 12:00 (1h)
[ETCE] Integrating ethical technologies and community engagement in multimedia technologies development
Amphitheater
› Addressing Demographic Bias in AI-Driven Driver Drowsiness Detection for Autonomous Vehicles
- Zenon Lamprou, Catalink Limited, University of Strathclyde [Glasgow] - Georgia Christodoulou, Catalink Limited
11:00-11:15 (15min)
› Designing and Integrating a Human-Centered Digital Co-Creation Toolkit for Ethical Multimedia and AI-Supported Participatory Processes[
- Peter Kafkias, DRAXIS Environmental SA
11:15-11:30 (15min)
› Generative AI for Urban Co-Design: A Survey and the SPICE Conceptual Framework for Participatory 3D Scene Editing
- Theodora Pistola, Centre for Research & Technology Hellas (CERTH) - Georgios Loupas, Centre for Research & Technology Hellas (CERTH) - Nefeli Georgakopoulou, Centre for Research & Technology Hellas (CERTH) - Ilias Koulalis, Centre for Research & Technology Hellas (CERTH)
11:30-11:45 (15min)
› From Voice to Space: Measuring Intersectional ASR Vulnerability in Inclusive Public Co-Design
- Azadeh Shojaei, Universitat Autònoma de Barcelona - Reyhaneh Sohrabi, Universitat Autònoma de Barcelona
11:45-11:53 (08min)
› Artificial Intelligence Driven Sign Language Interpreting Avatars
- Estella Oncins, Universitat Autònoma de Barcelona = Autonomous University of Barcelona = Universidad Autónoma de Barcelona - Pilar Orero, Universitat Autònoma de Barcelona = Autonomous University of Barcelona = Universidad Autónoma de Barcelona - Iris Serrat-Roozen, Universitat de València
11:53-12:00 (07min)
12:00 - 12:40 (40min)
Doctoral Symposium
Amphitheater
12:40 - 13:45 (1h05)
Lunch
13:45 - 14:15 (30min)
[ExFMA] Explainability and Fairness in Multimedia Analysis
Amphitheater
› A Content-Based Multimodal Approach for TV Series Recommendation: Tackling the Cold-Start Problem with Explainable Similarity Computation
- Camille Guinaudeau, Sciences et Technologies des Langues - LISN, Japanese French Laboratory of Informatics
13:45-14:00 (15min)
› Semantically Grounded Explainability for Pedestrian Detection Using Anatomical Proportions and Grid-Based Visualization
- Chiara Galdi, Eurecom [Sophia Antipolis]
14:00-14:15 (15min)
14:15 - 16:15 (2h)
Regular session 3
› AI-Based Thumbnail Generation for Soccer Clips using Multi-Signal Analysis
- Aliaan Azam, University of Oslo - Mehdi Houshmand Sarkhoosh, Forzasys
14:15-14:30 (15min)
› What Do Geolocation Models See? Object-Level Semantic Clusters as Country-Specific Evidence
- Emilie Durrieu, Laboratoire de recherche ENAC
14:30-14:45 (15min)
› Image Polysemy in Contrastive Vision-Language Learning
- Wangyuan Ding, Universiteit van Amsterdam
14:45-15:00 (15min)
› Knowledge-Driven Named Entity Recognition in French TV Video Transcripts
- Hoang Giang Vu, LIFAT - Delalandre Mathieu, LIFAT - Friburger Nathalie, LIFAT - Soulet Arnaud, LIFAT
15:00-15:15 (15min)
› Latent Adversarial Illusions: Defense-Aware Poisoning of Image Retrieval
- Minh-Huy Trinh, Ho Chi Minh City University of Science [Vietnam National University, HCM], Vietnam National University, Ho Chi Minh City / Đại học Quốc gia TP. Hồ Chí Minh
15:15-15:30 (15min)
› Learning Perceptual Hash Similarity for Image Copy Detection
- Maria Eirini Pegia, CERTH-ITI, Reykjavik University - Björn Þór Jónsson, Reykjavik University
15:30-15:45 (15min)
› Influence of Motion and Visual Content Properties on Depth Perception in Immersive Virtual Reality Environments
- Semihanur Aktay, TED University
15:45-15:53 (08min)
› FAISS-Indexed Vision Transformer Descriptors for Alzheimer's Disease Diagnosis
- Hadil Belanes, Innov'com, Sup'com, University of Carthage Ariana, Tunis, Tunisia, Synthèse et analyse d'images
15:53-16:00 (07min)
› Land Surface Irradiance Emulation with Gaussian Process Sampling and Neural Architecture Search
- Yasmine Karray, Centre d'études spatiales de la biosphère, Centre de Recherche en Numérique de Sfax, Université de Sfax - University of Sfax, Laboratory of Signals, systeMs, aRtificial Intelligence and neTworkS (SMARTS), Université de Toulouse
16:00-16:07 (07min)
16:00 - 16:20 (20min)
Coffee break
16:20 - 17:45 (1h25)
[VR4B] Video Retrieval for Beginners
› An Interactive Video Retrieval System for Novice Users with Example-Based Retrieval and Relevance Feedback
- Kazuya Ueki, Meisei University
16:20-17:40 (1h20)
› VIRA: An Interactive Video Retrieval Framework for Non-Expert Users
- Nhut Hoang Trang, Independent Researcher - Khanh-Vi Duong, Independent Researcher - Qui Ngoc Nguyen, University of Information Technology, Vietnam National University Ho Chi Minh City
16:20-17:40 (1h20)
› VISIONE at VR4B 2026: A Redesigned Interface for Interactive Video Retrieval
- Lucia Vadicamo, CNR Istituto di Scienza e Tecnologie dell'Informazione “A. Faedo” [Pisa]
16:20-17:40 (1h20)
› TapesVRy++: Guided Immersive Video Retrieval for Non-Expert Users
- Viet-Tham Huynh, Ho Chi Minh City University of Science [Vietnam National University, HCM] - Nhut-Thanh Le-Hinh, Ho Chi Minh City University of Science [Vietnam National University, HCM]
16:20-17:40 (1h20)
› Comparing Novice Interaction Across Two Video Retrieval Interfaces
- Goncalo Marcelino, Universiteit van Amsterdam - Omar Khan, IT University of Copenhagen
16:20-17:40 (1h20)
› EyeSpy4B: Saliency-Guided Video Retrieval Using Gaze for Novices
- Onanong Kongmeesub, Dublin City University [Dublin]
16:20-17:40 (1h20)
› SPOT: Spatially Prompted Object-Centric Text-Based Retrieval for Novice Video Search
- Rahel Arnold, Department of Mathematics and Computer Science, University of Basel - Raphael Waltenspül, Department of Mathematics and Computer Science, University of Basel
16:20-17:40 (1h20)
› Towards Usable Interactive Video Retrieval: A Multimodal System for Event Search
- Hai-Minh Phan, University of Science, VNU-HCM, Vietnam National University Ho Chi Minh
16:20-17:40 (1h20)
20:00 - 22:00 (2h)
Conference dinner
https://brasserielesbeauxarts.fr/
1 quai de la Daurade
31000 TOULOUSE
|
9:30 - 10:30 (1h)
[UHBER] Multimodal Data Analysis for Understanding of Human Behaviour, Emotions and their Reasons
Amphitheater
› Frogs, Sweets, and Heart Rates: Motivating Self-Reporting in Long-Term Workplace Pilots
- Elena Vildjiounaite, Freelance
09:30-09:45 (15min)
› Real-Time Assessment of Upper Limb Rehabilitation Exercise Using Discrete Fréchet distance
- (Poster) Mélanie FONTAINE, Laboratoire des technologies innovantes - UR UPJV 3899
09:30-10:30 (1h)
› Understanding Human Behavior in Open Surgical Skill Acquisition: A Multimodal Data Analysis of Cognitive Planning and Force Stability
- Noé Constans, Télécom SudParis, Gepromed
09:45-10:00 (15min)
› Controlling prostheses of upper limb amputees guided by intention of objects to grasp
- Ander Echezarreta, University of Bordeaux
10:00-10:07 (07min)
› Variational Contrastive Learning for Skeleton-based Action Recognition
- Dang Dinh NGUYEN, Telecom SudParis
10:07-10:15 (08min)
› From Pose to Behavior: Comparing Sequence Models for Detecting Stereotyped Motor Behaviors in Clinical Videos of Autistic Children
- Christian Lemler, Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy [Goethe Universität]
10:15-10:23 (08min)
10:30 - 11:00 (30min)
Regular session 4
Amphitheater
› Re-purposing Multimodal Large Language Models for Audio-Text Retrieval
- Jilan Xu, University of Oxford
10:30-10:45 (15min)
› Luggage re-identification: A study of strategies towards a real-world application
- Quentin Dhôte, Université Paris-Saclay, CEA, List, F-91120 Palaiseau, France
10:45-10:53 (08min)
› Multimodal Sleep Analytics for Athlete Performance: Chronotype Clustering, Readiness Prediction, and Anomaly Detection in Elite Female Football
- Michael Riegler, SimulaMet - Dag Johansen, UiT - Halvorsen Pål, SimulaMet
10:53-11:00 (07min)
11:00 - 11:20 (20min)
Coffee break
Main Hall
11:30 - 13:00 (1h30)
Regular session 5
Amphitheater
› Retrieval-Augmented Online Textual Memory for Episodic Memory Video Question Answering
- Raffaele Calì, Dipartimento di Matematica Informatica [University of Catania] - Giuseppe Lando, Dipartimento di Matematica Informatica [University of Catania] - Rosario Rosario Forte, Dipartimento di Matematica Informatica [University of Catania] - Antonino Furnari, Dipartimento di Matematica Informatica [University of Catania]
11:30-11:45 (15min)
› VideoHEDGE: Entropy-Based Hallucination Detection for Video-VLMs via Semantic Clustering and Spatiotemporal Perturbations
- Sushant Gautam, Simula Metropolitan Center for Digital Engineering AS [Oslo] - Cise Midoglu, Forzasys - Vajira Thambawita, Simula Metropolitan Center for Digital Engineering AS [Oslo] - Michael A. Riegler, Simula Metropolitan Center for Digital Engineering AS [Oslo] - Pål Halvorsen, Forzasys, Simula Metropolitan Center for Digital Engineering AS [Oslo]
11:45-12:00 (15min)
› User-Specific Multi-Modal Viewport Prediction in 360-Degree Videos
- Shini Koushal, Indian Institute of Technology [BHU Varanasi] - Dr. Bidyut Kr. Patra, Indian Institute of Technology [BHU Varanasi]
12:00-12:15 (15min)
› Leveraging Vision-Language Models for Personalized Activity and Object Identification in Lifelogging
- Manh Thai Van, Ho Chi Minh City University of Science [Vietnam National University, HCM]
12:15-12:23 (08min)
› On the Effect of Image Enhancement on Object Detection in Dual-Energy X-ray Images
- Anestis Anestis Kastellos, Centre for Research and Technology Hellas - Athanasios Psaltis, Centre for Research and Technology Hellas
12:23-12:30 (07min)
› Skeleton-Content Retrieval for Indian Sign Language: A Multi-Stream Prototype and Benchmark.
- Mohammad Rehan, Indian Institute of Technology Roorkee - Hazem Wannous, Centre for Digital Systems - Aditi Kar Gangopadhyay, Indian Institute of Technology Roorkee - Sugata Gangopadhyay, Indian Institute of Technology Roorkee
12:30-12:37 (07min)
› Spiking neural networks and interpersonal coordination of dancing dyads
- Théo Velletaz, EuroMov - Digital Health in Motion - IMT MINES Alès
12:37-12:45 (08min)
› Video-Based Assessment of Motor Skills in Parkinson's Disease Using Motion Representations from MotionGPT
- Vasiliki Georgali, Information Technologies Ιnstitute, Centre for Research and Technology Hellas - Kosmas Dimitropoulos, Information Technologies Ιnstitute, Centre for Research and Technology Hellas
12:45-12:53 (08min)
13:00 - 13:10 (10min)
Closing
Amphitheater
13:10 - 14:30 (1h20)
Lunch
14:30 - 17:00 (2h30)
Social event
Visits and other activities : Museum, Augustin, Aeroscopia
|
