posted by organizer: lukas_chr || 3897 views || tracked by 2 users: [display]

MuSe 2023 : The 4th International Multimodal Sentiment Analysis Challenge and Workshop @ ACM Multimedia 2023, Ottawa, Canada

FacebookTwitterLinkedInGoogle

Link: https://www.muse-challenge.org/
 
When Oct 29, 2023 - Nov 3, 2023
Where Ottawa, Canda
Submission Deadline Jul 7, 2023
Notification Due Jul 30, 2023
Final Version Due Aug 5, 2023
Categories    affective computing   machine learning   multimedia   artificial intelligence
 

Call For Papers

The Multimodal Sentiment Analysis Challenge (MuSe) 2023 is dedicated to multimodal sentiment and emotion recognition in different scenarios: https://www.muse-challenge.org

This year's edition features three different sub-challenges:


== Sub-Challenges, Datasets & Features ==

1. Emotional Mimicry Sub-challenge (MuSe-Mimic)
Predicting 3 discrete emotions (Approval, Uncertainty, Disappointment) from videos of mimicked emotional expressions. This sub-challenge utilises the novel Hume-Mimic dataset provided by Hume AI (https://www.hume.ai). It contains over 32 hours of audiovisual recordings of 557 different subjects, including transcripts.

2. Cross-Lingual Humor Detection Sub-challenge (MuSe-Humor)
Predicting the presence of humour in audio-visual recordings of football press conferences. In this sub-challenge, a cross-lingual setting is provided, where the training data comprises German recordings, while the test data is composed of English press conferences. Overall, more than 17 hours of data (audio, video, transcripts) are provided.

3. Personalisation Sub-challenge (MuSe-Personalisation)
Predicting emotional valence and psycho-physiological arousal signals from audio-visual recordings. In order to facilitate personalisation, parts of each test subject's labels are made available. MuSe-Personalisation uses the Ulm-TSST data set that includes audio, video, textual transcriptions and physiological signals (respiratory rate, ECG, BPM).

For the baseline paper see here: https://www.researchgate.net/publication/370100318_The_MuSe_2023_Multimodal_Sentiment_Analysis_Challenge_Mimicked_Emotions_Cross-Cultural_Humour_and_Personalisation

== How to Participate ==
Instructions are available at https://www.muse-challenge.org/challenge/participate. Data and features are available upon registration. Links to the baseline code and the preliminary baseline paper are also provided on the homepage (https://www.muse-challenge.org).


== Organisers ==
Shahin Amiriparian (University of Augsburg, GER)
Lukas Christ (University of Augsburg, GER)
Andreas König (University of Passau, GER)
Alan Cowen (Hume AI, USA)
Eva-Maria Meßner (University of Ulm, GER)
Erik Cambria (Nanyang Technological University/ SenticNet, SG)
Björn W. Schuller (Imperial College London/ audEERING, UK)

Related Resources

SUMAC 2025   The 7th ACM international workshop on analysis, understanding and promotion of heritage contents
AMLDS 2025   IEEE--2025 International Conference on Advanced Machine Learning and Data Science
Ei/Scopus-IPCML 2025   2025 International Conference on Image Processing, Communications and Machine Learning (IPCML 2025)
CBMI 2025   Content-Based Multimedia Indexing
ACM SAC 2025   40th ACM/SIGAPP Symposium On Applied Computing
IEEE- CCRIS 2025   2025 IEEE 6th International Conference on Control, Robotics and Intelligent System (CCRIS 2025)
MLMI 2025   2025 The 8th International Conference on Machine Learning and Machine Intelligence (MLMI 2025)
MuSE IC 2025   MuSE International Conference 2025: Multidisciplinary Sciences for Sustainability
SANER 2026   The 33rd IEEE International Conference on Software Analysis, Evolution and Reengineering
ASPLOS 2026   The ACM International Conference on Architectural Support for Programming Languages and Operating Systems - Summer