3e atelier international de l'ACM sur l'IA multimédia contre la désinformation (MAD'24)
Conférence internationale de l'ACM sur la récupération multimédia ICMR'24
Phuket, Thaïlande, 10-13 juin 2024
https://www.mad2024.aimultimedialab.ro/ https://easychair .org/my/conference?conf=mad2024 *** Appel à communications *** * Soumission des articles : 17 mars 2024 * Notification d'acceptation : 7 avril 2024 * Articles prêts à photographier : 25 avril 2024 * Atelier @ ACM ICMR 2024 : 10 juin 2024  La communication moderne ne repose plus uniquement sur les médias classiques comme les journaux ou la télévision, mais s'effectue plutôt sur les réseaux sociaux, en temps réel et avec des interactions en direct entre les utilisateurs. Toutefois, l’accélération de la quantité d’informations disponibles a également conduit à une augmentation de la quantité et de la qualité des contenus trompeurs, de la désinformation et de la propagande. A l'inverse, la lutte contre la désinformation, à laquelle participent quotidiennement agences de presse et ONG (entre autres) pour éviter les risques de distorsion de l'opinion des citoyens, est devenue encore plus cruciale et exigeante, notamment pour ce qui concerne des sujets sensibles comme la politique. , la santé et la religion. Les campagnes de désinformation exploitent, entre autres, des outils basés sur l'IA pour la génération et la modification de contenu : des contenus visuels, vocaux, textuels et vidéo hyperréalistes ont émergé sous le nom collectif de « deepfakes », et plus récemment avec l'utilisation de grands modèles linguistiques. (LLM) et les grands modèles multimodaux (LMM), minant la crédibilité perçue du contenu médiatique. Il est donc encore plus crucial de contrer ces avancées en concevant de nouveaux outils d’analyse capables de détecter la présence de contenus synthétiques et manipulés, accessibles aux journalistes et aux vérificateurs de faits, robustes et fiables, et éventuellement basés sur l’IA pour atteindre de plus grandes performances. Les futures recherches multimédias sur la détection de la désinformation reposent sur la combinaison de différentes modalités et sur l’adoption des dernières avancées en matière d’approches et d’architectures d’apprentissage profond. Cela soulève de nouveaux défis et questions qui doivent être résolus afin de réduire les effets des campagnes de désinformation. L'atelier, dans sa troisième édition, accueille les contributions liées à différents aspects de la détection, de l'analyse et de l'atténuation de la désinformation basée sur l'IA. Les sujets d'intérêt comprennent, sans s'y limiter : - Détection de la désinformation dans le contenu multimédia (par exemple, vidéo, audio, textes, images) - Méthodes de vérification multimodales - Détection des médias synthétiques et manipulés - Forensique multimédia - Diffusion et effets de la désinformation dans les médias sociaux - Analyse des campagnes de désinformation dans des domaines socialement sensibles - Robustesse de la vérification des médias contre les attaques contradictoires et les complexités du monde réel   


























- Équité et non-discrimination de la détection de la désinformation dans le contenu multimédia
- Expliquer les technologies de désinformation/détection de désinformation aux utilisateurs non experts
- Aspects temporels et culturels de la désinformation
- Partage et gouvernance des ensembles de données dans l'IA pour la désinformation
- Ensembles de données pour la détection de la désinformation et la vérification multimédia
- Ouvert ressources, par exemple, ensembles de données, outils logiciels
- Grands modèles linguistiques pour analyser et atténuer les campagnes de désinformation
- Grands modèles multimodaux pour la vérification des médias
- Systèmes et applications de vérification multimédia
- Techniques de fusion, d'assemblage et de fusion tardive de systèmes
- Cadres d'analyse comparative et d'évaluation


*** Directives de soumission ***

Lors de la préparation de votre soumission, veuillez respecter strictement les instructions de l'ACM ICMR 2024, afin de garantir la pertinence du processus d'examen et l'inclusion dans les actes de la bibliothèque numérique de l'ACM. Les instructions sont disponibles ici :  https://mad2024.aimultimedialab.ro/submissions/ .



*** Comité d'organisation ***

Cristian Stanciu, Université Politehnica de Bucarest, Roumanie
Luca Cuccovillo, Fraunhofer IDMT, Allemagne
Bogdan Ionescu, Université Politehnica de Bucarest, Roumanie
Giorgos Kordopatis-Zilos, Université technique tchèque de Prague, Tchéquie
Symeon Papadopoulos, Centre pour Research and Technology Hellas, Thessalonique, Grèce
Adrian Popescu, CEA LIST, Saclay, France
Roberto Caldelli, CNIT et Mercatorum University, Italie

L'atelier est soutenu dans le cadre du projet H2020 AI4Media - A European Excellence Centre for Media, Society and Democracy ( https:/ /www.ai4media.eu/ ), le projet Horizon Europe  vera.ai  - VERification Assisted by Artificial Intelligence ( https://www.veraai.eu/ ) et le projet Horizon Europe AI4Debunk - Outils participatifs d'assistance basés sur l'IA pour soutenir une ligne de confiance Activité des citoyens et démystification de la désinformation ( https://ai4debunk.eu/ ).


Au nom des organisateurs,

Cristian Stanciu
https://www.aimultimedialab.ro/

Back

Top

Here is a link to join us remotely: https://univ-montp3-fr.zoom.us/j/95371655397?pwd=C0hjVtrbDcTdhmbTuHd1Q1cWS9njHe.1
Meeting ID: 953 7165 5397
Passcode: 974510

Back

Top

The Technical University of Madrid (UPM) will once more organize the 'Madrid UPM Machine Learning and Advanced Statistics' summer school. The summer school will be held in Boadilla del Monte, near Madrid, from June 17th to June 28th. This year's edition comprises 12 week-long courses (15 lecture hours each), given during two weeks (six courses each week). Attendees may register in each course independently. No restrictions, besides those imposed by timetables, apply on the number or choice of courses.

We would like to remind you that early registration for the Madrid UPM Machine Learning and Advanced Statistics summer school is open until June 2nd (included). The summer school will be held in Boadilla del Monte, near Madrid, from June 19th to June 30th. This year's edition comprises 12 week-long courses (15 lecture hours each), given during two weeks (six courses each week). Attendees may register in each course independently. No restrictions, besides those imposed by timetables, apply on the number or choice of courses.

Early registration is now *OPEN*. Extended information on course programmes, price, venue, accommodation and transport is available at the school's website:

http://www.dia.fi.upm.es/MLAS

There is a 25% discount for members of Spanish AEPIA and SEIO societies. 

Please, forward this information to your colleagues, students, and whomever you think may find it interesting.

Best regards,

Pedro Larrañaga, Concha Bielza, Bojan Mihaljević and Laura Gonzalez Veiga.
-- School coordinators.

*** List of courses and brief description ***

* Week 1 (June 17th - June 23rd, 2024) *

1st session: 9:45-12:45
Course 1: Bayesian Networks (15 h)
      Basics of Bayesian networks. Inference in Bayesian networks. Learning Bayesian networks from data. Real applications. Practical demonstration: R.

Course 2: Time Series(15 h)
      Basic concepts in time series. Linear models for time series. Time series clustering. Practical demonstration: R.
     
2nd session: 13:45-16:45
Course 3: Supervised Classification (15 h)
      Introduction. Assessing the performance of supervised classification algorithms. Preprocessing. Classification techniques. Combining multiple classifiers. Comparing supervised classification algorithms. Practical demonstration: python.

Course 4: Statistical Inference (15 h)
      Introduction. Some basic statistical tests. Multiple testing. Introduction to bootstrap methods. Introduction to Robust Statistics. Practical demonstration: R. 

3rd session: 17:00 - 20:00
Course 5: Deep Learning (15 h)
      Introduction. Learning algorithms. Learning in deep networks. Deep Learning for Computer Vision. Deep Learning for Language. Practical session: Python notebooks with Google Colab with keras, Pytorch and Hugging Face Transformers.

Course 6: Bayesian Inference (15 h)
      Introduction: Bayesian basics. Conjugate models. MCMC and other simulation methods. Regression and Hierarchical models. Model selection. Practical demonstration: R and WinBugs.
     

* Week 2 (June 26th - June 28th, 2024) *

1st session: 9:45-12:45

Course 7: Feature Subset Selection (15 h)
      Introduction. Filter approaches. Embedded methods. Wrapper methods. Additional topics. Practical session: R and python.

Course 8: Clustering (15 h)
      Introduction to clustering. Data exploration and preparation. Prototype-based clustering. Density-based clustering. Graph-based clustering. Cluster evaluation. Miscellanea. Conclusions and final advice. Practical session: R.

2nd session: 13:45-16:45
Course 9: Gaussian Processes and Bayesian Optimization (15 h)
      Introduction to Gaussian processes. Sparse Gaussian processes. Deep Gaussian processes. Introduction to Bayesian optimization. Bayesian optimization in complex scenarios. Practical demonstration: python using GPytorch and BOTorch.
     
Course 10: Explainable Machine Learning (15 h)
      Introduction. Inherently interpretable models. Post-hoc interpretation of black box models. Basics of causal inference. Beyond tabular and i.i.d. data. Other topics. Practical demonstration: Python with Google Colab.
         
3rd session: 17:00-20:00
Course 11:  SVMs, Kernel Methods and Regularized Learning (15 h)
      Regularized learning. Kernel methods. SVM models. SVM learning algorithms. Practical session: Python Anaconda with scikit-learn.
     
Course 12: Hidden Markov Models (15 h)
      Introduction. Discrete Hidden Markov Models. Basic algorithms for Hidden Markov Models. Semicontinuous Hidden Markov Models. Continuous Hidden Markov Models. Unit selection and clustering. Speaker and Environment Adaptation for HMMs. Other applications of HMMs. Practical session: HTK.

Back

Top

                                            Colloque international

           « Nouvelles Perspectives d’analyse musicale de la voix »

              Université Lumière Lyon 2, Lyon, 20-21 juin 2024

                                    APPEL À COMMUNICATIONS

Thématiques suggérées (liste non-limitative) :

• Analyse structurelle de la voix chantée ou du parlé musicalisé.

• Techniques d'analyse harmonique et mélodique appliquées à la voix.

• Méthodes et techniques d’analyse de la voix.

• Nouvelles perspectives technologiques et computationnelles d'analyse de la voix.

• Approches stylistiques ou rhétoriques dans l'analyse de la voix.

• Exploration acoustique, physiologique et interdisciplinaire de techniques vocales spécifiques, d’effets interprétatifs ou de modalités variées d’utilisation de la voix.

• Étude du rythme, du timbre vocal, du phrasé, etc.

Modalités de soumission : Nous vous invitons à soume8re votre proposi=on de communica=on avant le 1ER FÉVRIER 2024. Les propositions, qui devront comporter un résumé (2500 signes maximum, en français ou en anglais) et une courte notice bio-bibliographique, seront à faire parvenir conjointement à Antoine Petit (antoine.petit@univ-lyon2.fr) et Céline Chabot-Canet (celine.chabot-canet@univ-lyon2.fr). Les réponses seront communiquées au plus tard le 8 février 2023. Ce colloque donnera lieu à une publication des actes. Comité scientifique : Céline Chabot-Canet, Muriel Joubert, Antoine Petit, Axel Roebel, Catherine Rudent. Comité d’organisation : Antoine Petit (doctorant), Céline Chabot-Canet (MCF), Passages Arts & Li8ératures (XX-XXI), Université Lumière Lyon 2. Dans le cadre du projet ANR « Analyse et tRansformation du Style de chant » (ANR-19-CE38-0001-03).

Back

Top

Call for Abstracts for the workshop 'Prosodic features of language learners' fluency'

https://l2fluency.lst.uni-saarland.de/

This workshop is a satellite event of 'Speech Prosody' to be held in Leiden (The Netherlands) on 1st of July, 2024. Its aim is to bring together colleagues from two research communities to focus on speech fluency: spoken second/foreign language (L2) on the one hand and speech prosody on the other.

In the past, fluency was often ignored in speech prosody research (as reflected in the Handbook of Language Prosody (2022) and also in the Speech Prosody conferences). Moreover, fluency and timing are only rarely treated together with intonation-related aspects in L2 research. However, a broader ranging view on L2 sentence prosody would be beneficial to the construction of theories concerning the acquisition of L2 prosody and applications such as assessments in teaching, exercises for individual learning, assessments and automatic testing of spoken performances. Likewise, research of language learning does not seem to be very much integrated into speech prosody research. This concerns both theoretical and methodological aspects but also acquisition and annotation of learner data, e.g. in learner corpora.

Thus, the scope of the workshop includes topics like measuring fluency, assessment of fluency (human experts, non-experts, and machines), learner corpora and annotation of disfluencies, elements and combinations of disfluencies (e.g. filler particles, disfluent pauses, lengthenings, repetitions, repairs), varying degrees of fluency in different speech styles and tasks, fluency and L2 proficiency levels, intonational aspects of fluency, visual aspects of fluency (e.g. hand-arm gestures, eye-gazing, torso movement), teaching methods for fluency improvement in L2 speech production and perception.

Keynote speakers are Lieke van Maastricht (Radboud University Nijmegen) and Malte Belz (Humboldt University Berlin).

Interested colleagues are invited to submit a two-page abstract (first page for text, second page for illustrations, tables, and references) to be reviewed by an expert committee. Only oral presentations are planned. In addition to this workshop, we are discussing the possibility of editing a special (open) issue in a recognised journal (e.g. 'Journal of Second Language Pronunciation' or 'Studies in Second Language Acquisition') to which we would encourage presenters of workshop papers to contribute.

Important dates: abstract submission deadline: 8 April, notification of acceptance: 1 May, workshop day: 1 July 2024.

Organisers: Jürgen Trouvain, Bernd Möbius (both Saarland University) and Nivja de Jong (Leiden University)

Back

Top

Speech Prosody Workshop Announcement

CROSSIN: Intonation at the Crossroads

Speech Prosody Satellite Workshop, Leiden, Saturday 6 July 2024

WORKSHOP ANNOUNCEMENT AND CALL FOR POSTER PRESENTATIONS

Intonation is studied by different disciplines in which the research focus varies. One element these approaches have in common is that they must all address intonation meaning. This applies whether researchers are mostly interested in the phonological representation of intonation, its interaction with syntax, semantics, and pragmatics, or its role in communication and speech processing. These perspectives complement each other, yet it is often the case that research focusing on one does not give full consideration to the others: for instance, syntactic approaches to the role of intonation in expressing focus may overlook differences in phonological form in focus expression, while pragmatic approaches may assume that each meaning nuance is directly expressed by a different tune; conversely, studies on intonation phonetics and phonology do not always fully consider meaning. 

The aim of this workshop is to reach a more comprehensive view, by bringing together researchers working on intonation from different perspectives so they can enter into dialogue with and learn from each another. The main questions of the workshop are:

1. What is the relationship between syntax, semantics, pragmatics, and intonation? Can we expect a one-to-one correspondence between intonation categories or tunes, on the one hand, and focus or other semantic or pragmatic functions, on the other?
2. How can we best understand and model intonation meaning and intonation’s role in conversation and processing?

We invite abstracts addressing the questions above. The selected abstracts will be presented in a poster session. If there is sufficient interest, poster presentations will be published as a special issue or collection.

Keynote speakers: The workshop also includes invited talks by Stavros Skopeteas (Göttingen), Anja Arnhold (Alberta), and commentaries by James German (Aix-Marseille) and Claire Beyssade (Paris 8). The workshop will end with a general round-table discussion. For more information on the workshop, visit https://www.sprintproject.io/crossinworkshop  or http://tinyurl.com/y7zj8h5f .

Important dates: abstract submission deadline: 31 March; notification of acceptance: 30 April; workshop day: 6 July 2024

Abstract Guidelines

Abstracts should be written in English and should present original research not already submitted to Speech Prosody. The  text should not exceed one A4 page , though an additional page for references, examples, and figures may also be added. The following formatting conventions apply: Times New Roman font, size 12, 2.54 cm (1 inch) margins, single spacing. Submissions should be sent as anonymized pdf files to sprintonation@gmail.com by 31 March 2024 at 24:00 AoE. Please provide author details in your email.

Organizers: Amalia Arvaniti, Stella Gryllia, Jiseung Kim, Riccardo Orrico, Alanna Tibbs (Radboud University)

Back

Top

Conférence JEP-TALN-2024

Du 8 au 12 juillet 2024

Toulouse, France

======================

Les équipes de recherche SAMoVA, MELODI et IRIS de l’Institut de Recherche en Informatique de Toulouse (IRIT, UMR 5505), l’équipe PLC du laboratoire Cognition, Langues, Langage, Ergonomie (CLLE, UMR 5263) et l’axe neurocognition langagière, linguistique et phonétique cliniques du laboratoire de NeuroPsychoLinguistique (LNPL, URI EA 4156) organisent conjointement à Toulouse les 35ème Journées d’Études sur la Parole (JEP), la 31ème Conférence sur le Traitement Automatique des Langues Naturelles (TALN) et la 26ème Rencontre des Étudiants Chercheurs en Informatique pour le Traitement Automatique des Langues, dénommée (RECITAL).

https://jep-taln2024.sciencesconf.org/

----------------------------------

Dates importantes (JEP-TALN-RECITAL) :

-   Soumission des articles : *** février 2024 (date définitive) ***

-   Notification aux auteurs : 25 avril 2024

-   Date de la conférence : 8 au 12 juillet 2024

- Proposition d atelier : ***22 février 2024 (date définitive) ***

Les thématiques de la conférence s’inscrivent dans les catégories suivantes, sans y être limitées pour autant.

TALN-RECITAL

-   Phonétique, phonologie, morphologie, étiquetage morphosyntaxique

-   Syntaxe, grammaires, analyse syntaxique, chunking

-   Sémantique, pragmatique, discours

-   Sémantique lexicale et distributionnelle

-   Aspects linguistiques et psycholinguistiques du TAL

-   Ressources pour le TAL

-   Méthodes d’évaluation pour le TAL

-   Applications du TAL (recherche et extraction d’information, question-réponse, traduction, génération, résumé, dialogue, analyse d’opinions, simplification, etc.)

-   TAL et multimodalité (parole, vision, etc.)

-   TAL et multilinguisme

-   TAL pour le Web et les réseaux sociaux

-   TAL et langues peu dotées

-   TAL et langue des signes

-   Implications sociales et éthiques du TAL

-   TAL et linguistique de corpus

-   TAL et Humanités numériques

JEP

-   Acoustique de la parole

-   Acquisition de la parole et du langage

-   Analyse, codage et compression de la parole

-   Applications à composantes orales (dialogue, indexation, etc)

-   Apprentissage d’une langue seconde

-   Communication multimodale

-   Dialectologie

-   Évaluation, corpus et ressources

-   Langues en danger

-   Modèles de langage

-   Parole audio-visuelle

-   Pathologies de la parole

-   Phonétique et phonologie

-   Phonétique clinique

-   Production / Perception de la parole

-   Prosodie

-   Psycholinguistique

-   Reconnaissance et compréhension de la parole

-   Reconnaissance de la langue

-   Reconnaissance du locuteur

-   Signaux sociaux, sociophonétique

-   Synthèse de la parole

Le nombre de pages des soumissions pour JEP/TALN/RECITAL est libre, mais compris entre 6 et 10 pages (selon le détail de l’appel et hors références/annexes). Le principe est que la taille de la soumission doit être cohérente avec son contenu. Les relecteurs jugeront un article sur sa qualité et cette adéquation.

Les feuilles de style et le détail des appels sont disponibles sur le site web de la conférence : https://jep-taln2024.sciencesconf.org/

Lien de soumission https://easychair.org/conferences/?conf=jeptaln2024

Back

Top

Appel à ateliers de JEPTALN 2024

Conférence JEPTALN 2024

8 - 12 juillet 2024

Dans le cadre des conférences conjointes JEPTALN2024, nous sollicitons des propositions d'ateliers. Les ateliers doivent porter sur une thématique particulière de traitement automatique des langues ou de la parole afin de rassembler quelques exposés plus ciblés que lors des conférences plénières.

Chaque atelier a son propre président et son propre comité de programme. Le responsable de l'atelier est chargé de la communication sur celui-ci, de l'appel à soumissions et de la coordination de son comité de programme.

Les organisateurs de JEPTALN2024 s'occuperont de la logistique (e.g. gestion des salles, pauses café et diffusion des articles).

Les ateliers auront lieu en parallèle durant une journée ou une demi-journée (2 à 4 sessions de 1h30) le lundi 8 juillet 2024 sur le campus de l’Université Jean Jaurès de Toulouse.

Dates importantes

-   Date limite de soumission des propositions d'atelier : 15 février 2024

-   Réponse du comité de programme : 29 février 2024

Modalités de proposition

Les propositions d'ateliers (1 à 2 pages A4 en format PDF) comprendront :

-   le nom et l'acronyme de l’atelier

-   une description synthétique du thème de l'atelier

-   le comité d'organisation

-   le comité scientifique provisoire ou pressenti

-   l'adresse du site web

-   la durée souhaitée pour la réalisation de l'atelier (1 journée

  ou 1/2 journée) et l'audience potentielle

Les propositions d'ateliers devront être envoyées sous forme électronique à jose.moreno@irit.fr et julie.mauclair@irit.fr avec pour entête de courriel : [Atelier JEP TALN 2024].

Modalités de sélection

Les propositions d'atelier seront examinées par des membres des comités de programme de JEP, TALN, par l’AFCP et le CPERM de l'ATALA. Les critères suivants seront considérés pour acceptation :

-   l'adéquation aux thèmes de l'une ou l'autre des conférences

-   l'originalité de la proposition

Format

Les conférences auront lieu en français (ou en anglais pour les non-francophones). Les articles soumis devront suivre le format de JEPTALN 2024 (nombre de pages à la discrétion du comité de programme de l'atelier). La soumission des versions finales devra suivre le calendrier de la conférence principale.

Back

Top

Call for Papers: 7th Laughter and Other Non-Verbal Vocalisations
> Workshop - July 16-17 2024
>
>
> We are excited to announce the 7th Laughter and Other Non-Verbal
> Vocalisations Workshop (bit.ly/LaughterWorkshop2024) on July 16-17 at
> Queen’s University Belfast. The workshop will be a pre-conference
> event, part of the 2024 Conference of the International Society for
> Research on Emotion (www.isre2024.org).
>
> Non-verbal vocalisations in human-human and human-machine interactions
> play important roles in displaying social and affective behaviours and
> in managing the flow of interaction. Laughter, sighs, clicks, filled
> pauses, and short utterances such as feedback responses are among some
> of the non-verbal vocalisations that are being increasingly studied
> from various research fields. However, much is still unknown about the
> phonetic or visual characteristics of non-verbal vocalisations
> (production/encoding), their relations to the social actions they are
> part of, their perceived meanings (perception/decoding), and their
> ordering in interaction. Furthermore, with the increased interest for
> more naturalness in human-machine interaction, current times also
> invite exploring how these phenomena can be integrated in speech
> applications.
>
> Research themesinclude, but are not restricted to, these aspects of
> laughter and other non-verbal vocalisations:
>
>  *
>
>    Articulation, acoustics, and perception
>
>  *
>
>    Interaction and pragmatics
>
>  *
>
>    Affective and evaluative meanings
>
>  *
>
>    Social perception and organisation
>
>  *
>
>    Disfluency
>
>  *
>
>    Technology applications
>
> Researchers are invited to submit extended abstracts(2 pages long,
> including figures and references) describing their work, including work
> in progress. The deadline for submission is March 15th, 2024. More
> information about the submission process can be found on our website
> (bit.ly/LaughterWorkshop2024).
>
> There will be twokeynote presentations on the topics treated by the
> workshop, delivered by Prof. Carolyn McGettigan (University College
> London, UK) and Prof. Margaret Zellers (Kiel University, Germany).
>
> Looking forward to receiving your contributions and welcoming you at
> the workshop in July!

Back

Top

13th International Conference

on Voice Physiology and Biomechanics

Erlangen, Germany 22nd-26th of July 2024

we cordially invite you to participate in the 13th International Conference on Voice Physiology and Biomechanics, July 22nd – 26th of 2024!

After the successful hosting in 2012, we are pleased to welcome you back in Erlangen, Germany! There will be two days of workshops prior to the three days of conference and several social events in the beautiful Nuremberg Metropolitan Region.

The workshops (July 22nd-23rd) and the conference (July 24th-26th) will focus on voice physiology and biomechanics including computational, numerical and experimental modeling, machine learning, tissue engineering, laryngeal pathologies and many more. Abstract submission and registration will be open from November 1st, 2023.

We are looking forward to your contributions and to seeing you in Erlangen, July 2024!

Back

Top

The 7th IEEE International Conference on
Multimedia Information Processing and Retrieval (MIPR 2024)

August 7 – 9, 2024 San Jose, CA, USA

http://www.ieee-mipr.org
https://sites.google.com/view/mipr2024

Joint conference collocation with the IEEE International Conference on
Information Reuse and Integration for Data Science (IRI) 2024

A vast amount of multimedia data is becoming accessible, making the
understanding of spatial and/or temporal phenomena crucial for many
applications. This necessitates the utilization of techniques in the
processing, analysis, search, mining, and management of multimedia
data. The 7th IEEE International Conference on Multimedia Information
Processing and Retrieval (IEEE-MIPR 2024) will take place in San Jose,
CA, USA on August 7–9, 2024, to provide a forum for original research
contributions and practical system design, implementation, and
applications of multimedia information processing and retrieval. The
target audiences include university researchers, scientists, industry
professionals, software engineers and graduate students. The event
includes a main conference as well as multiple associated keynote
speeches, workshops, challenge contests, tutorials, and panels.

Topics

Generative and Foundation Models in Multimedia
- AI-generated Media
- Foundation Models in Vision
- Security of Large AI Models
- Multimodal Media Detection
- Generation and Detection with Diffusion Models
- Media Generation with Large Language Models
- Visual and Vision-Language Pre-training
- Generic Vision Interface
- Alignments in Text-to-image Generation
- Large Multimodal Models
- Multimodal Agents

Trustworthy AI in Multimedia
- AI Reliability for Multimedia Applications and Systems
- AI Fairness for Multimedia Applications and Systems
- AI Robustness for Multimedia Applications and Systems
- Attack and Defense for Multimedia Applications and Systems

Video/Audio in Multimedia
- Speech/Voice Synthesis
- Analysis of Conversation
- Speaker and Language Identification
- Audio Signal Analysis
- Spoken Language Generation
- Automatic Speech Recognition
- Spoken Dialogue and Conversational AI Systems

Vision and Content Understanding
- Multimedia Telepresence and Virtual/Augmented/Mixed Reality
- Visual Concept Detection
- Object Detection and Tracking
- 3D Modeling, Reconstruction, and Interactive Applications
- Multimodal/Multisensor Interfaces, Integration, and Analysis
- Effective and Scalable Solution for Big Data Integration
- Affective and Perceptual Multimedia

Multimedia Retrieval
- Multimedia Search and Recommendation
- Web-Scale Retrieval
- Relevance Feedback, Active/Transfer Learning
- 3D and Sensor Data Retrieval
- Multimodal Media (Images, Videos, Texts, Graph/Relationship) Retrieval
- High-Level Semantic Multimedia Features

Machine/Deep Learning/Data Mining
- Deep Learning in Multimedia Data and Multimodal Fusion
- Deep Cross-Learning for Novel Features and Feature Selection
- High-Performance Deep Learning (Theories and Infrastructures)
- Spatio-Temporal Data Mining
- Novel Dataset for Learning and Multimedia

Multimedia Systems and Infrastructures
- Multimedia Systems and Middleware
- Software Infrastructure for Data Analytics
- Distributed Multimedia Systems and Cloud Computing

Networking in Multimedia
- Internet Scale System Design
- Information Coding for Content Delivery

Data Management
- Multimedia Data Collection, Modeling, Indexing, or Storage
- Data Integrity, Security, Protection, Privacy
- Standards and Policies for Data Management

Novel Applications
- Multimedia Applications for Health and Sports
- Multimedia Applications for Culture and Education
- Multimedia Applications for Fashion and Living
- Multimedia Applications for Security and Safety

Internet of Multimedia Things
- Real-Time Data Processing
- Autonomous Systems (Driverless Cars, Robots, Drones, etc.)
- Mobile and Wearable Multimedia

User Experience and Engagement
- Quality of Experience
- User Engagement
- Emotional and Social Signals

Paper Submission: The conference will accept regular papers (6 pages),
short papers (4 pages), and demo papers (4 pages), including
references. Authors are encouraged to compare their approaches,
qualitatively or quantitatively, with existing work and explain the
strengths and weaknesses of the new approaches. The CMT online
submission site is at https://cmt3.research.microsoft.com/MIPR2024.
All accepted papers presented in MIPR 2024 will be published in the
conference proceedings which will also be available online at the IEEE
Xplore digital library. 

Important Dates:
- Paper (regular/short/demo) submission: April 15, 2024 Pacific Time
- Paper review available: May 8, 2024
- Notification of acceptance: May 20, 2024
- Camera-ready deadline: June 17, 2024

Back

Top

ASVspoof 5 challenge (Robust Speech Deepfake Detection and Automatic Speaker Verification) 

The registration for participation in

is now open: https://www.asvspoof.org/

The registration deadline is July 10th, 2024

The challenge features two tracks:

Track 1: Speech deepfake detection (DF) - 'real vs fake' speech detection

Track 2: Spoofing-robust automatic speaker verification (SASV)

ASVspoof is a community-driven, not-for-profit challenge series which promotes the development and benchmarking of generalizable speech deepfake detection and automatic speaker verification systems intended to operate reliably in the face of spoofing attacks. The challenge data is constructed using public speech resources, and organizers provide baseline systems and reference metrics. Compared to previous challenge editions, ASVspoof 5 involves a substantially larger amount of data, enabling participants to develop more sophisticated detection models. To promote robustness, as well as the development of solutions with practical applications in the wild, ASVspoof 5 focuses on non-studio-quality speech data.

How to participate?

1. Read the evaluation plan, available at www.asvspoof.org 

2. Join the e-mail list: send an e-mail to sympa@lists.asvspoof.org with 'subscribe ASVspoof5' as the subject line.

3. Register: https://shorturl.at/cqrtK 

Timeline:

- Training and development data available:      May 20, 2024

- Challenge leaderboard (Codalab) opens:        June 05, 2024 

- Evaluation data available:                  June 12, 2024 

- Challenge submissions due:                  July 17, 2024

- ASVspoof 5 workshop paper deadline:           July 31, 2024

- Acceptance notifications:                   August 10, 2024

- ASVspoof 5 workshop at Interspeech:         August 31, 2024

Please note that access to Codalab will be granted to registered participants only. Further details are available in the evaluation plan, which will be supplemented with additional details as the challenge progresses.

The ASVspoof 5 challenge organisers

info@asvspoof.org

Back

Top

La parole devient un moyen de plus en plus important pour l'interaction homme-machine avec de nombreux déploiements dans les domaines de la biométrie, de la médecine légale et, surtout, de l'accès à l'information via des assistants vocaux virtuels. Parallèlement à ces développements, le besoin d'algorithmes et d'applications robustes et sécurisés qui protègent la sécurité et la confidentialité de l'utilisateur est apparu à l'avant-garde de la recherche et du développement basés sur la parole.  

La quatrième édition du Symposium sur la sécurité et la confidentialité dans la communication vocale, combinée cette année au  VoicePrivacy Challenge , se concentre sur la parole et la voix à travers lesquelles nous nous exprimons. Étant donné que la communication vocale peut être utilisée pour commander à des assistants virtuels de transporter des émotions ou de s'identifier, le symposium tente de répondre à la question de savoir comment renforcer la sécurité et la confidentialité des types de représentation vocale dans une interaction homme/machine centrée sur l'utilisateur. Le symposium constate donc que les échanges interdisciplinaires sont très demandés et vise à rassembler des chercheurs et des praticiens de plusieurs disciplines, plus précisément : le traitement du signal, la cryptographie, la sécurité, l'interaction homme-machine, le droit et l'anthropologie.

L'initiative VoicePrivacy est à la tête des efforts visant à développer des solutions de préservation de la confidentialité pour la technologie vocale. Il vise à consolider la communauté nouvellement formée pour développer la tâche et les mesures et évaluer les progrès réalisés dans les solutions d'anonymisation à l'aide d'ensembles de données, de protocoles et de mesures communs. VoicePrivacy prend la forme d’un défi compétitif. Conformément aux éditions précédentes du VoicePrivacy Challenge, l'édition actuelle se concentre sur l'anonymisation de la voix. Les participants doivent développer des systèmes d'anonymisation pour supprimer l'identité du locuteur tout en gardant intacts le contenu et les attributs paralinguistiques. Cette édition se concentre sur la préservation de l’état émotionnel, qui constitue l’attribut paralinguistique clé dans de nombreuses applications réelles de l’anonymisation vocale. Tous les participants sont encouragés à soumettre au symposium SPSC des articles liés à leur participation au défi, ainsi que d'autres articles scientifiques liés à l'anonymisation des locuteurs et à la confidentialité de la voix. Plus de détails peuvent être trouvés sur la page Web du VoicePrivacy Challenge :  https://www.voiceprivacychallenge.org/ . 

Afin de renforcer les efforts pour les deux événements, faciliter les discussions communes et étendre les échanges interdisciplinaires, nous avons décidé de regrouper nos équipes et d'organiser un événement commun. Pour le colloque général, nous acceptons les contributions sur des sujets connexes, ainsi que les rapports d'avancement, la diffusion de projets ou les discussions théoriques et les « travaux en cours ». En outre, les invités du monde universitaire, de l'industrie et des institutions publiques ainsi que les étudiants intéressés sont invités à assister à la conférence sans avoir à apporter leur propre contribution. Toutes les soumissions acceptées apparaîtront dans les actes du symposium publiés dans les archives ISCA.

Soumission:

Les articles destinés au symposium SPSC doivent contenir jusqu'à huit pages de texte. La durée doit être choisie de manière appropriée pour présenter le sujet à une communauté interdisciplinaire. Les soumissions d'articles doivent être conformes au format défini dans les directives de préparation des articles et tel que détaillé dans le  kit de l'auteur . Les articles doivent être soumis via le système de soumission d'articles en ligne via le lien sur le   site Web du SPSC . La langue de travail de la conférence est l'anglais et les articles doivent être rédigés en anglais. Tous les articles acceptés seront publiés dans les archives ISCA aux côtés des articles Interspeech et des ateliers ISCA associés.

Commentaires: 

Au moins trois examens en double aveugle seront effectués et nous visons à obtenir les commentaires d'experts interdisciplinaires pour chaque soumission. Pour les contributions au VoicePrivacy Challenge, l’examen se concentrera sur les descriptions et les résultats des systèmes. 

Rendez-vous importants:

Lieu: 

Le lieu du Symposium sera publié prochainement, nous prévoyons de le faire co-localiser avec Interspeech 2024. Une participation hybride est possible

Back

Top

Call for Labs Proposals @CLEF 2024

At its 25th edition, the Conference and Labs of the Evaluation Forum (CLEF) is a continuation of the very successful series of evaluation campaigns of the Cross Language Evaluation Forum (CLEF) which ran between 2000 and 2009, and established a framework of systematic evaluation of information access systems, primarily through experimentation on shared tasks. As a leading annual international conference, CLEF uniquely combines evaluation laboratories and workshops with research presentations, panels, posters and demo sessions. In 2024, CLEF takes place in September,  9-12 at the University of Grenoble Alpes, France.

Researchers and practitioners from all areas of information access and related communities are invited to submit proposals for running evaluation labs as part of CLEF 2024. Proposals will be reviewed by a lab selection committee, composed of researchers with extensive experience in evaluating information retrieval and extraction systems. Organisers of selected proposals will be invited to include their lab in the CLEF 2024 labs programme, possibly subject to suggested modifications to their proposal to better suit the CLEF lab workflow or timeline.

Background

The CLEF Initiative (http://www.clef-initiative.eu/) is a self-organised body whose main mission is to promote research, innovation, and development of information access systems with an emphasis on multilingual information in different modalities - including text and multimedia - with various levels of structure. CLEF promotes research and development by providing an infrastructure for:

1. independent evaluation of information access systems;

2. investigation of the use of unstructured, semi-structured, highly-structured, and semantically enriched data in information access; 

3. creation of reusable test collections for benchmarking; 

4. exploration of new evaluation methodologies and innovative ways of using experimental data; 

5. discussion of results, comparison of approaches, exchange of ideas, and transfer of knowledge.
   
   

Scope of CLEF Labs

We invite submission of proposals for two types of labs:

1. “Campaign-style” Evaluation Labs for specific information access problems (during the twelve months period preceding the conference), similar in nature to the traditional CLEF campaign “tracks”. Topics covered by campaign-style labs can be inspired by any information access-related domain or task.
   
   

2. Labs that follow a more classical “workshop” pattern, exploring evaluation methodology, metrics, processes, etc. in information access and closely related fields, such as natural language processing, machine translation, and human-computer interaction.
   
   

We highly recommend organisers new to the CLEF format of shared task evaluation campaigns to first consider organising a lab workshop to discuss the format of their proposed task, the problem space and practicalities of the shared task. The CLEF 2024 programme will reserve about half of the conference schedule for lab sessions. During the conference, the lab organisers will present their overall results in overview presentations during the plenary scientific paper sessions to give non-participants insights into where the research frontiers are moving. During the conference, lab organisers are expected to organise separate sessions for their lab with ample time for general discussion and engagement with all participants - not just those presenting campaign results and papers. Organisers should plan time in their sessions for activities such as panels, demos, poster sessions, etc. as appropriate. CLEF is always interested in receiving and facilitating innovative lab proposals. 

Potential task proposers unsure of the suitability of their task proposal or its format for inclusion at CLEF are encouraged to contact the CLEF 2024 Lab Organizing Committee Chairs to discuss its suitability or design at an early stage.

Proposal Submission

Lab proposals must provide sufficient information to judge the relevance, timeliness, scientific quality, benefits for the research community, and the competence of the proposers to coordinate the lab. Each lab proposal should identify one or more organisers as responsible for ensuring the timely execution of the lab. Proposals should be 3 to 4 pages long and should provide the following information:

1. Title of the proposed lab.
    

2. A brief description of the lab topic and goals, its relevance to CLEF and the significance for the field.
    

3. A brief and clear statement on usage scenarios and domain to which the activity is intended to contribute, including the evaluation setup and metrics.
    

4. Details on the lab organiser(s), including identifying the task chair(s) responsible for ensuring the running of the task. This should include details of any previous involvement in organising or participating in evaluation tasks at CLEF or similar campaigns.
    

5. The planned format of the lab, i.e., campaign-style (“track”) or workshop.
    

6. Is the lab a continuation of an activity from previous year(s) or a new activity?  

1. For activities continued from previous year(s): Statistics from previous years (number of participants/runs for each task), a clear statement on why another edition is needed, an explicit listing of the changes proposed, and a discussion of lessons to be learned or insights to be made.
   
   

2. For new activities: A statement on why a new evaluation campaign is needed and how the community would benefit from the activity.
    

7. Details of the expected target audience, i.e., who do you expect to participate in the task(s), and how do you propose to reach them.
    

8. Brief details of tasks to be carried out in the lab. The proposal should clearly motivate the need for each of the proposed tasks and provide evidence of its capability of attracting enough participation. The dataset which will be adopted by the Lab needs to be described and motivated in the perspective of the goals of the Labs; also indications on how the dataset will be shared are useful. It is fine for a lab to have a single task, but labs often contain multiple closely related tasks, needing a strong motivation for more than 3 tasks, to avoid useless fragmentation.
    

9. Expected length of the lab session at the conference: half-day, one day, two days. This should include high-level details of planned structure of the session, e.g. participant presentations, invited speaker(s), panels, etc., to justify the requested session length.
    

10. Arrangements for the organisation of the lab campaign: who will be responsible for activities within the task; how will data be acquired or created, what tools or methods will be used, e.g., how will necessary queries be created or relevance assessment carried out; any other information which is relevant to the conduct of your lab.
     

11. If the lab proposes to set up a steering committee to oversee and advise its activities, include names, addresses, and homepage links of people you propose to be involved.
    
    

Lab proposals must be submitted at the following address:

https://easychair.org/conferences/?conf=clef2024

choosing the “CLEF 2024 Lab Proposals” track.

Reviewing Process

Each submitted proposal will be reviewed by the CLEF 2024 Lab Organizing Committee. The acceptance decision will be sent by email to the responsible organiser by 28 July 2023. The final length of the lab session at the conference will be determined based on the overall organisation of the conference and the number of participant submissions received by a lab.

Advertising Labs at CLEF 2023 and ECIR 2024

Organisers of accepted labs are expected to advertise their labs at both CLEF 2023 (18-21 September 2023, Thessaloniki, Greece) and ECIR 2024 (24-28 March 2024, Glasgow, Scotland). So, at least one lab representative should attend these events.

Advertising at CLEF 2023 will consist of displaying a poster describing the new lab, running a break-out session to discuss the lab with prospective participants, and advertising/announcing it during the closing session.

Advertising at ECIR 2024 will consist of submitting a lab description to be included in ECIR 2024 proceedings (11 October 2023) and advertising the lab in a booster session during ECIR 2024.

Mentorship Program for Lab Proposals from newcomers

CLEF 2019 introduced a mentorship program to support the preparation of lab proposals for newcomers to CLEF. The program will be continued at CLEF 2024 and we encourage newcomers to refer to Friedberg et al. (2015) for initial guidance on preparing their proposal:

Friedberg I, Wass MN, Mooney SD, Radivojac P. Ten simple rules for a community computational challenge. PLoS Comput Biol. 2015 Apr 23;11(4):e1004150.

The CLEF newcomers mentoring program offers help, guidance, and feedback on the writing of your draft lab proposal by assigning a mentor to you, who help you in preparing and maturing the lab proposal for submission. If your lab proposal falls into the scope of an already existing CLEF lab, the mentor will help you to get in touch with those lab organisers and team up forces.

Lab proposals for mentorship must be submitted at the following address:

https://easychair.org/conferences/?conf=clef2024

choosing the “CLEF 2024 Lab Mentorship” track.

Important Dates

• 29 May 2023: Requests for mentorship submission (only newcomers)

• 29 May 2023 - 16 June 2023: Mentorship period

• 7 July 2023: Lab proposals submission (newcomers and veterans)

• 28 July 2023: Notification of lab acceptance

• 18-21 Sep 2023: Advertising Accepted Labs at CLEF 2023, Thessaloniki, Greece

• 11 October 2023: Submission of short lab description for ECIR 2024

• 13 November 2023: Lab registration opens

• 24-28 March 2024: Advertising labs at ECIR 2024, Glasgow, UK

CLEF 2024 Lab Chairs

• Petra Galuscakova, University of Stavanger, Norway

• Alba García Seco de Herrera, University of Essex, UK

CLEF 2024 Lab Mentorship Chair

• Liana Ermakova, Université de Bretagne Occidentale, France

• Florina Piroi, TU Wien, Austria

Back

Top

Call for Demonstrations for the 21st International Conference on Content-based Multimedia Indexing (CBMI)

September 18 – 20, 2024 in Reykjavík, Iceland

Conference website: https://cbmi2024.org/

(Apologies if you receive multiple copies of this call)

The 21st International Conference on Content-based Call for Demonstrations for the 21st International Conference on Content-based Multimedia Indexing (CBMI)

September 18 – 20, 2024 in Reykjavík, Iceland Multimedia Indexing (CBMI) welcomes the submission of demonstration papers. We invite authors to report on and showcase novel and compelling demonstrations (software, methods and experiences) in all topic areas relevant to CBMI. 

Submission Guidelines

The length of the papers should be up to 4 pages, in IEEE conference format, plus 1 page for references. One or two additional page(s) should be appended to illustrate what the demo involves and how it will be conducted on-site. This additional content will not be published in the conference proceedings, should the submission be accepted!  If possible, we also invite you to include a URL linking to a short video (max. 3 min) that shows the demonstration in action. 

Demonstration papers are subject to peer review in a single-blind process according to criteria such as novelty, interestingness, applications of or enhancements to state-of-the-art, and potential impact.

Submission Deadline

The extended submission deadline is May 6th, 2024 (AoIE). To submit your paper, follow the instructions in the submission guidelines. 

Infrastructure on Site

The conference will provide a table, power outlet, screen, wireless (shared) internet and a poster board. Presenters are expected to bring the necessary equipment (computers, etc.) themselves. If you have special needs (e.g., more space), please include a related note in the appendix of your submission (“Special Needs” section).

Should you have any questions regarding submissions, please contact the chairs at demo-chairs@cbmi2024.org. 

Back

Top

Call for Papers: Special Session on 'Multimedia Indexing for eXtended Reality' at CBMI 2024

https://cbmi2024.org/?page_id=100#MmIXR

21st International Conference on Content-based Multimedia Indexing (CBMI 2024).
18-20 September 2024, Reykjavik, Iceland - https://cbmi2024.org/

DESCRIPTION:
Extended Reality (XR) applications rely not only on computer vision for navigation and object placement but also require a range of multimodal methods to understand the scene or assign semantics to objects being captured and reconstructed. Multimedia indexing for XR thus encompasses methods for processes during XR authoring, such as indexing content to be used for scene and object reconstruction, as well as during the immersive experience, such as object detection and scene segmentation.
The intrinsic multimodality of XR applications involves new challenges like the analysis of egocentric data (video, depth, gaze, head/hand motion) and their interplay. XR is also applied in diverse domains, e.g., manufacturing, medicine, education, and entertainment, each with distinct requirements and data. Thus, multimedia indexing methods must be capable of adapting to the relevant semantics of the particular application domain.

TOPICS OF INTEREST:

• Multimedia analysis for media mining, adaptation (to scene requirements), and description for use in XR experiences (including but not limited to AI-based approaches)

• Processing of egocentric multimedia datasets and streams for XR (e.g., egocentric video and gaze analysis, active object detection, video diarization/summarization/captioning)

• Cross- and multi-modal integration of XR modalities (video, depth, audio, gaze, hand/head movements, etc.)

• Approaches for adapting multimedia analysis and indexing methods to new application domains (e.g., open-world/open-vocabulary recognition/detection/segmentation, few-shot learning)

• Large-scale analysis and retrieval of 3D asset collections (e.g., objects, scenes, avatars, motion capture recordings)

• Multimodal datasets for scene understanding for XR

• Generative AI and foundation models for multimedia indexing and/or synthetic data generation

• Combining synthetic and real data for improving scene understanding

• Optimized multimedia content processing for real-time and low-latency XR applications

• Privacy and security aspects and mitigations for XR multimedia content
  
   

IMPORTANT DATES:
Submission of papers: 22 March 2024
Notification of acceptance: 3 June 2024
CBMI conference: 18-20 September 2024

SUBMISSION:
The session will be organized as an oral presentation session. The contributions to this session will be long papers describing novel methods or their adaptation to specific applications or short papers describing emerging work or open challenges.

SPECIAL SESSION ORGANISERS:
Fabio Carrara, Artificial Intelligence for Multimedia and Humanities Laboratory, ISTI-CNR, Pisa, Italy

Werner Bailer, Intelligent Vision Applications Group, JOANNEUM RESEARCH, Graz, Austria

Lyndon J. B. Nixon, MODUL Technology GmbH and Applied Data Science School at MODUL University, Vienna, Austria

Vasileios Mezaris, Information Technologies Institute / Centre for Research and Technology Hellas, Thessaloniki, Greece

Back

Top

Call for Papers: Special Session on 'Multimodal Insights for Disaster Risk Management and Applications (MIDRA)' at CBMI 2024

https://cbmi2024.org/?page_id=100#MIDRA

21st International Conference on Content-based Multimedia Indexing (CBMI 2024).
18-20 September 2024, Reykjavik, Iceland - https://cbmi2024.org/

Disaster management in all its phases from preparedness, prevention, response, and recovery is in abundance of multimedia data, including valuable assets like satellite images, videos from UAVs or static cameras, and social media streams. The value of such multimedia data for operational purposes in disaster management is not only useful for civil protection agencies but also for the private sector that quantifies risk. Indexing data from crisis events presents Big Data challenges due to its variety, velocity, volume and veracity for effective analysis and retrieval.

The advent of deep learning and multimodal data fusion offers an unprecedented opportunity to overcome these challenges and fully unlock the potential of disaster event multimedia data. Through the strategic utilization of different data modalities, researchers can significantly enhance the value of these datasets, uncovering insights that were previously beyond reach, giving actionable information and supporting real-life decision-making procedures.

This special session actively seeks research papers in the domain of multimodal analytics and their applications in the context of crisis event monitoring through knowledge extraction and multimedia understanding. Emphasis is placed on recognizing the intrinsic value of spatial information when integrated with other data modalities.

The special session serves as a collaborative platform for communities focused on specific crisis events, such as forest fires, volcano unrest or eruption, earthquakes, floods, tsunamis and extreme weather events, which have increased significantly due to the climate crisis in our era. It fosters the exchange of ideas, methodologies, and software tailored to address challenges in these domains, aiming to encourage fruitful collaborations and the mutual enrichment of insights and expertise among diverse communities.

This special session includes presentation of novel research within the following domains:

• Lifelog computing
• Urban computing
• Satellite computing and earth observation
• Multimodal data fusion
• Social media

Within these domains, the topics of interest include (but are not restricted to):

• Multimodal analytics and retrieval techniques for crisis event multimedia data.
• Deep learning and neural networks for interpretability, understanding, and explainability in artificial intelligence applied to natural disasters.
• Satellite image analysis and fusion with in-situ data for crisis management.
• Integration of multimodal data for comprehensive risk assessment.
• Application of deep learning techniques to derive insights for risk mitigation.
• Development of interpretative models for better understanding of risk factors.
• Utilization of diverse data modalities (text, images, sensors) for risk management.
• Implementation of multimodal analytics in predicting and managing natural disasters.
• Application of multimodal insights in insurance risk assessment.
• Enhanced decision-making through the fusion of geospatial and multimedia data.

Important Dates:
Submission of papers: 22 March 2024
Notification of acceptance: 3 June 2024
CBMI conference: 18-20 September 2024

Organisers:

• Maria Pegia, Information Technologies Institute / Centre for Research and Technology Hellas, Greece.
• Ilias Gialampoukidis, Information Technologies Institute / Centre for Research and Technology Hellas, Greece.
• Ioannis Papoutsis, National Observatory of Athens & National Technical University of Athens, Greece.
• Krishna Chandramouli, Venaka Treleaf GbR, Germany.
• Stefanos Vrochidis, Information Technologies Institute / Centre for Research and Technology Hellas, Greece.

Please direct correspondence to midra@cbmi2024.org

Back

Top

The 21st International Conference on Content-based Multimedia Indexing (CBMI 2024) will be held in Reykjavik, Iceland next September 18-20: https://cbmi2024.org/

The conference will bring together leading experts from academia and industry interested in the broad field of content-based multimedia indexing and applications.

The Special Session on 'Explainability in Multimedia Analysis' (ExMA), addresses the analysis of multimedia applications, such as person detection/tracking, face recognition or lifelog analysis, which may affect sensitive personal information. This raises both legal issues, e.g. concerning data protection and regulations in the ongoing European AI regulation, as well as ethical issues, related to potential bias in the system or misuse of these technologies. This special session focuses on AI-based explainability technologies in multimedia analysis.

The conference CBMI’2024 is supported by ACM SIGMM and the proceedings will be available at ACM Digital Library.

We would like to invite you to consider contributing a paper to this special session.

CBMI's important dates: https://cbmi2024.org/?page_id=211

Looking forward to see you at CBMI 2024.
With best regards,
Chiara Galdi

Special session organisers: Chiara Galdi, Martin Winter, Romain Giot, Romain Bourqui

Back

Top

The 21st International Conference on Content-based Multimedia Indexing (CBMI 2024) takes place September 18-20 in Reykjavik, Iceland.

We are delighted to have, as part of the conference, a Special Session on Audio entitled: Content based Indexing for audio and music: from analysis to synthesis 

Abstract: Audio has long been a key component of multimedia research. As far as indexing is concerned, the research and industrial context has changed drastically in the last 20 years or so. Today, applications of audio indexing range from karaoke applications to singing voice synthesis and creative audio design. This special session aims at bringing together researchers that aim at proposing new tools or paradigms to investigate audio and music processing in the context of indexation and corpus-based generation.

You are kindly encouraged to submit a paper related to the topic of the special session according to the CBMI guidelines : 

• Regular full papers: 6 pages, plus additional pages for the list of references

• Regular short papers: 4 pages, plus additional pages for the list of references

Important dates: 

• March 22: Regular and special session paper submissions

• June 3: Notification of acceptance 

• Early July: Camera ready version of accepted papers

As of now, we already have 3 invited talks addressing the following topics : 

• Cynthia C. S. Liem, Doğa Taşcılar, and Andrew M. Demetriou A quest through interconnected datasets: lessons from highly-cited ICASSP papers

• Rémi Mignot, Geoffroy Peeters Learning invariance to sound modifications for music indexing and alignment

• Cyrus Vahidi Large-scale music indexing for multimodal similarity search

Please join us in Reykjavik !!

Kindly yours,

François Pachet and Mathieu Lagrange

contact us: mathieu lagrange ls2n fr

Back

Top

—————— 
Björn Þór Jónsson (bjorn@ru.is)
Professor
Department of Computer Science
Reykjavik University (http://www.ru.is/)
Iceland

Back

Top

Sixth International Workshop on the History of Speech Communication Research

September 20–21, 2024, Budapest

After highly popular sessions at ICPhS in Prague this year and an exceptional workshop „Lacerda 120” in Porto last year, we are happy to announce that the next HSCR workshop will take place in Budapest next year on Sept 20 and 21, organised by Judit Bóna and Mária Gósy of the Department of Applied Linguistics and Phonetics of ELTE University. The manuscript submission deadline is May 15, 2024. All details can be found at the workshop website: https://hscr2024.elte.hu/

The aim of this workshop is to bring scholars together who study the history of speech science to learn more on the methods, findings and results of our predecessors and to better understand the speech research community’s present achievements.

Speech has been investigated from different perspectives, which necessitates a range of approaches and scientific methods. Previous contributions analyzed the contextual background of individual researchers, investigated how specific research practices developed over time, examined the various kinds of approach of researchers to their material and the link between the form and the meaning in speech communication research.

The special focus of the 6th HSCR workshop will be on the development of the specific fields of speech communication, such as emerging phonology, progression in analysis of both speech sounds and prosody, speech technology, growing body of psycholinguistics, sociophonetics and clinical phonetics, etc. Researchers are encouraged to mine deep into history to find the early steps and advancement of these specific fields of speech communication. The knowledge of our predecessors is frequently unknown, forgotten or ignored for several reasons, and thus the past attainments are not appropriately integrated in our common consciousness regarding speech science.

As always, contributions on other topics from the history of speech communication research will also be welcome. The unfolded facts of the phonetic endeavor in the history of speech science may heavily inspire the present research.

Manuscripts should be sent to the email address of the workshop: hscr2024@gmail.com. Please, use the templates for your paper.

The proceedings will be published in the book series Studientexte zur Sprachkommunikation at TUDpress (Technical University Dresden). The HSCR proceedings will be published in print and also stored electronically in the ISCA archive.

For any inquiries, please use the workshop email address: hscr2024@gmail.com

Back

Top

Nous organisons la 13e édition du Colloque des Jeunes Chercheurs de Praxiling (UMR 5267) qui aura lieu à Montpellier du 17 au 18 octobre 2024. La thématique du colloque est la suivante : « Vulnérabilité et langage : langues, locuteurs, discours ». 

Ce colloque s’adresse aux jeunes chercheurs s’intéressant à la thématique de la vulnérabilité sous des angles divers : langues vulnérables, locuteurs vulnérabilisés, discours sur la vulnérabilité émanant ou non des locuteurs en situation de vulnérabilité. 

Vous trouverez ci-joint l’appel à communication contenant l’argumentaire ainsi que toutes les conditions de rédaction et modalités de participation. La date de clôture de l’appel est fixée au 15 juin 2024 et la proposition est à envoyer à l’adresse suivante : cjc.praxiling.2024@gmail.com.

Vous retrouverez toutes les informations sur le site : https://cjc-praxiling2024.www.univ-montp3.fr

Bien cordialement, 

Le comité d’organisation : Lou BRUN, Myriam CASALONE, Elora DANJEAN, Ahamada KASSIME – Praxiling UMR 5267 Université Paul-Valéry Montpellier 3

Back

Top

7th International Workshop on Multimedia Content Analysis in Sports (MMSports'24) @ ACM Multimedia, Oct 28 – Nov 1, 2024, Melbourne, Australia

We'd like to invite you to submit your paper proposals for the 7th International Workshop on Multimedia Content Analysis in Sports to be held in Melbourne, Australia together with ACM Multimedia 2024. The ambition of this workshop is to bring together researchers and practitioners from many different disciplines to share ideas and methods on current multimedia/multimodal content analysis research in sports. We welcome multimodal-based research contributions as well as best-practice contributions focusing on the following (and similar, but not limited to) topics:

- annotation and indexing in sports 

- tracking people/ athlete and objects in sports

- activity recognition, classification, and evaluation in sports

- 3D scene and motion reconstruction in sports

- event detection and indexing in sports

- performance assessment in sports

- injury analysis and prevention in sports

- data driven analysis in sports

- graphical augmentation and visualization in sports

- automated training assistance in sports

- camera pose and motion tracking in sports

- brave new ideas / extraordinary multimodal solutions in sports

- personal virtual (home) trainers/coaches in sports

- datasets in sports

- graphical effects in sports

- alternative sensing in sports (beyond the visible spectrum)

- multimodal perception in sports

- exploiting physical knowledge in learning systems for sports

- sports knowledge discovery

- narrative generation and narrative analysis in sports

- mobile sports application

- multimedia in sports beyond video, including 3D data and sensor data

Submissions can be of varying length from 4 to 8 pages, plus additional pages for the reference pages. There is no distinction between long and short papers, but the authors may themselves decide on the appropriate length of their paper. All papers will undergo the same review process and review period.

Please refer to the workshop website for further information: 

http://mmsports.multimedia-computing.de/mmsports2024/index.html

IMPORTANT DATES

Submission Due:                           19 July 2024 

Acceptance Notification:             5 August 2024

Camera Ready Submission:         19 August 2024 

Workshop Date:                            TBA; either Oct 28 or Nov 1, 2024

ACM MMSports’24 Chairs: Thomas Moeslund, Rainer Lienhart and Hideo Saito

Back

Top

https://icmi.acm.org/2024/call-for-papers/

https://new.precisionconference.com/submissions/icmi24a

Important Dates

Novelty will be assessed along two dimensions: scientific novelty and technical novelty. Accepted papers at ICMI 2024 will need to be novel along one of the two dimensions:

• Scientific Novelty: Papers should bring new scientific knowledge about human social interactions, including human-computer interactions. For example, discovering new behavioral markers that are predictive of mental health or how new behavioral patterns relate to children’s interactions during learning. It is the responsibility of the authors to perform a proper literature review and clearly discuss the novelty in the scientific discoveries made in their paper.
• Technical Novelty: Papers should propose novelty in their computational approach for recognizing, generating or modeling multimodal data. Examples include: novelty in the learning and prediction algorithms, in the neural architecture, or in the data representation. Novelty can also be associated with new usages of an existing approach.

Commitment to ethical conduct is required and submissions must adhere to ethical standards in particular when human-derived data are employed. Authors are encouraged to read the ACM Code of Ethics and Professional Conduct (https://ethics.acm.org/).

The theme of this year’s ICMI conference revolves around “Equitability and environmental sustainability in multimodal interaction technologies.” The focus is on exploring how multimodal systems and multimodal interactive applications can serve as tools to bridge the digital divide, particularly in underserved communities and countries, with a specific emphasis on those in Latin America and the Caribbean. The conference aims to delve into the design principles that can render multimodal systems more equitable and sustainable in applications such as health and education, thereby catalyzing positive transformations in development for historically marginalized groups, including racial/ethnic minorities and indigenous peoples. Moreover, there is a crucial exploration of the intersection between multimodal interaction technologies and environmental sustainability. This involves examining how these technologies can be crafted to comprehend, disseminate, and mitigate the adverse impacts of climate change, especially in the Latin America and Caribbean region. The conference endeavors to explore the potential of multimodal systems in fostering community resilience, raising awareness, and facilitating education related to climate change, thereby contributing to a holistic approach that encompasses both social and environmental dimensions.

Additional topics of interest include but are not limited to:

• Affective computing and interaction
• Cognitive modeling and multimodal interaction
• Gesture, touch and haptics
• Healthcare, assistive technologies
• Human communication dynamics
• Human-robot/agent multimodal interaction
• Human-centered A.I. and ethics
• Interaction with smart environment
• Machine learning for multimodal interaction
• Mobile multimodal systems
• Multimodal behaviour generation
• Multimodal datasets and validation
• Multimodal dialogue modeling
• Multimodal fusion and representation
• Multimodal interactive applications
• Novel multimodal datasets
• Speech behaviours in social interaction
• System components and multimodal platforms
• Visual behaviours in social interaction
• Virtual/augmented reality and multimodal interaction

Back

Top

*******************************************************

SPECOM-2024 – FIRST CALL FOR PAPERS

*******************************************************

26th International Conference on Speech and Computer (SPECOM-2024)

November 25-28, 2024

Crowne Plaza hotel, Belgrade, Serbia

Web: https://specom2024.ftn.uns.ac.rs/

ORGANIZERS

The conference SPECOM-2024 is organized by the Faculty of Technical Sciences University of Novi Sad and the School of Electrical Engineering University of Belgrade in cooperation with the Telecommunications Society of Serbia

FOUNDERS

SPECOM series was founded by St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences (SPIIRAS) of the St. Petersburg Federal Research Center of the Russian Academy of Sciences (SPC RAS)

CONFERENCE TOPICS

SPECOM attracts researchers, linguists and engineers working in the following areas of speech science, speech technology, natural language processing, human-computer interaction:

• Affective computing

• Audio-visual speech processing

• Corpus linguistics

• Computational paralinguistics

• Deep learning for audio processing

• Feature extraction

• Forensic speech investigations

• Human-machine interaction

• Language identification

• Large language models

• Multichannel signal processing

• Multilingual speech technology

• Multimedia processing

• Multimodal analysis and synthesis

• Natural language generation

• Natural language understanding

• Sign language processing

• Speaker diarization

• Speaker identification and verification

• Speech and language resources

• Speech analytics and audio mining

• Speech and voice disorders

• Speech-based applications

• Speech driving systems in robotics

• Speech enhancement

• Speech perception

• Speech recognition and understanding

• Speech synthesis

• Speech translation systems

• Spoken dialogue systems

• Spoken language processing

• Text mining and sentiment analysis

• Virtual and augmented reality

• Voice assistants

SATELLITE EVENTS

26th International Conference SPECOM will be organized together with the 32nd Telecommunications Forum TELFOR-2024: https://www.telfor.rs/en/

OFFICIAL LANGUAGE

The official language of the event is English. However, papers on processing of languages other than English are strongly encouraged.

FORMAT OF THE CONFERENCE

The conference program will include presentation of invited talks, oral presentations, and poster/demonstration sessions.

SUBMISSION OF PAPERS

Authors are invited to submit full papers of 10-15 pages formatted in the Springer LNCS style. Each paper will be reviewed by at least three independent reviewers (single-blind), and accepted papers will be presented either orally or as posters. Papers submitted to SPECOM must not be under review by any other conference or publication during the SPECOM review cycle, and must not be previously published or accepted for publication elsewhere. The authors are invited to submit their papers using the on-line submission system: https://easychair.org/conferences/?conf=specom2024

DEADLINES

July 01, 2024 ....................... Submission of full papers

September 03, 2024 ........... Notification of acceptance/rejection

September 15, 2024 ........... Camera-ready papers

October 01, 2024 ................ Early registration

PROCEEDINGS

SPECOM Proceedings will be published by Springer as a book in the Lecture Notes in Artificial Intelligence (LNAI/LNCS) series listed in all major international citation databases.

GENERAL CHAIRS

Vlado DELIĆ – Faculty of Technical Sciences University of Novi Sad, Novi Sad, Serbia

Alexey KARPOV – SPIIRAS, SPC RAS, St. Petersburg, Russia

CONTACTS

All correspondence regarding the conference should be addressed to SPECOM-2024 Secretariat

E-mail: specom2024@uns.ac.rs

Web: https://specom2024.ftn.uns.ac.rs

Back

Top

The 2nd International Conference on Foundation and Large Language Models (FLLM2024)

Hybrid Event

https://fllm2024.fllm-conference.org/index.php

26-29 November, 2024 | Dubai, UAE

Technically Co-Sponsored by IEEE UAE Section

FLLM 2024 CFP:

With the emergence of foundation models (FMs) and Large Language Models (LLMs) that are trained on large amounts of data at scale and adaptable to a wide range of downstream applications, Artificial intelligence is experiencing a paradigm revolution. BERT, T5, ChatGPT, GPT-4, Falcon 180B, Codex, DALL-E, Whisper, and CLIP are now the foundation for new applications ranging from computer vision to protein sequence study and from speech recognition to coding. Earlier models had a reputation of starting from scratch with each new challenge. The capacity to experiment with, examine, and comprehend the capabilities and potentials of next-generation FMs is critical to undertaking this research and guiding its path. Nevertheless, these models are currently inaccessible as the resources required to train these models are highly concentrated in industry, and even the assets (data, code) required to replicate their training are frequently not released due to their demand in the real-time industry. At the moment, mostly large tech companies such as OpenAI, Google, Facebook, and Baidu can afford to construct FMs and LLMS. Despite the expected widely publicized use of FMs and LLMS, we still lack a comprehensive knowledge of how they operate, why they underperform, and what they are even capable of because of their emerging global qualities. To deal with these problems, we believe that much critical research on FMs and LLMS would necessitate extensive multidisciplinary collaboration, given their essentially social and technical structure.

The International Conference on Foundation and Large Language Models (FLLM) addresses the architectures, applications, challenges, approaches, and future directions. We invite the submission of original papers on all topics related to FLLMs, with special interest in but not limited to:

•     Architectures and Systems
• Transformers and Attention
• Bidirectional Encoding
• Autoregressive Models
• Massive GPU Systems
• Prompt Engineering
• Multimodal LLMs
• Fine-tuning
•     Challenges
• Hallucination
• Cost of Creation and Training
• Energy and Sustainability Issues
• Integration
• Safety and Trustworthiness
• Interpretability
• Fairness
• Social Impact
•     Future Directions
• Generative AI
• Explainability and EXplainable AI
• Retrieval Augmented Generation (RAG)
• Federated Learning for FLLM
• Large Language Models Fine-Tuning on Graphs
• Data Augmentation
•     Natural Language Processing Applications
• Generation
• Summarization
• Rewrite
• Search
• Question Answering
• Language Comprehension and Complex Reasoning
• Clustering and Classification
•     Applications
• Natural Language Processing
• Communication Systems
• Security and Privacy
• Image Processing and Computer Vision
• Life Sciences
• Financial Systems

Submissions Guidelines and Proceedings

Manuscripts should be prepared in 10-point font using the IEEE 8.5' x 11' two-column format. All papers should be in PDF format, and submitted electronically at Paper Submission Link. A full paper can be up to 8 pages (including all figures, tables and references). Submitted papers must present original unpublished research that is not currently under review for any other conference or journal. Papers not following these guidelines may be rejected without review. Also submissions received after the due date, exceeding length limit, or not appropriately structured may also not be considered. Authors may contact the Program Chair for further information or clarification. All submissions are peer-reviewed by at least three reviewers. Accepted papers will appear in the FLLM Proceeding, and be published by the IEEE Computer Society Conference Publishing Services and be submitted to IEEE Xplore for inclusion. Submitted papers must include original work, and must not be under consideration for another conference or journal. Submission of regular papers up to 8 pages and must follow the IEEE paper format. Please include up to 7 keywords, complete postal and email address, and fax and phone numbers of the corresponding author. Authors of accepted papers are expected to present their work at the conference. Submitted papers that are deemed of good quality but that could not be accepted as regular papers will be accepted as short papers.

Important Dates:

• Paper submission deadline: June 30, 2024
• Notification of acceptance: September 15, 2024
• Camera-ready Submission: October 10, 2024

Contact:

Please send any inquiry on FLLM to: info@fllm-conference.org

Back

Top

We are pleased to extend an invitation to you to participate in the upcoming Fearless Steps APOLLO Workshop. Our workshop delves into exploring speech communication, technology, and the extensive audio of the historic NASA Apollo program.

The Fearless Steps APOLLO Community Resource, supported by NSF, is a unique and massive naturalistic communications resource. This resource, derived from the Apollo missions, offers a rare glimpse into team-based problem-solving in high-stakes environments, with a rich variety of speech and language data providing invaluable data for researchers, scientists, historians, and technologists.

The Fearless Steps APOLLO corpus contains 30 time-synchronized channels, which capture all NASA Apollo team communications. The PAO (Public Affairs Officer) channel reflects all live public broadcast TV/radio contexts streamed by NASA during the missions. This channel is similar to all Broadcast news corpora.

Our workshop aims to showcase featured speakers, panel discussions, and present the latest findings in speech and language processing. We will explore facets of the Fearless Steps APOLLO corpus, the largest publicly available naturalistic team-based historical audio and meta-data resource.

Topics Covered:

We will be exploring several key areas, including:

1. Big Data Recovery and Deployment in the Fearless Steps APOLLO initiative.

2. Applications in Education, History, and Archival efforts.

3. Insights into Communication Science and Psychology, particularly in Group Dynamics and Team Cohesion.

4. Speech and Language Technology (SLT) development, including ASR, SAD, speaker recognition, and conversational topic detection. 

Workshop Structure:

1. Discuss advancements in digitizing Apollo audio and machine learning solutions for audio diarization.

2. Explore team communication dynamics through speech processing.

3. Explore the utility of Fearless Steps APOLLO resource for: SpchTech (Speech & Language Technology), CommSciPsychTeam (Communication Sciences & Team-based Psychology), & EducArchHist (Education, History, & Archival) communities.

4. The FEARLESS STEPS Challenge, a community engagement and data generation initiative.

The workshop will feature oral talks, including an overview of Fearless Steps APOLLO resource, including Team presentations on systems evaluated for the Fearless Steps Challenge dataset.

Instructions for Authors:

We invite authors to submit a short 1-page research overview that involves the Fearless Steps APOLLO resource. Please submit your Abstracts through our dedicated portal.

The workshop format will include oral presentations for accepted abstracts, which will be announced after the paper submission. Submissions in the form of 1-page abstracts ( and an optional additional page for references, figures, or preliminary results) are encouraged. Detailed formatting instructions and sample PDFs are available on our website. The Complete Fearless Steps Challenge (Phase-1 to Phase-4) Corpora & Naturalistic (Apollo-11 & Apollo-13) corpora can be accessed by filling out a short survey form here: FS-APOLLO Corpora Download Access

The deadline for workshop Abstract submission is set for March 1, 2024. We will announce the acceptance of the Abstracts on March 15, 2024. Both in-person and remote participation options will be available, with a focus on fostering a collaborative environment. Papers accepted to ICASSP 2024 are welcome as Abstract submissions, as well as original research following our format guidelines.

We believe this workshop will be a pivotal step in advancing speech technology and research. We look forward to your participation in enriching the potential of the Apollo Resource and inspiring new approaches in collaborative problem-solving.

For more details, please visit our workshop website.

Back

Top