ICMI'23 CALL FOR MULTIMODAL GRAND CHALLENGES

TK
Truong, Khiet (UT-EEMCS)
Wed, Jan 11, 2023 10:34 AM

ICMI'23 CALL FOR MULTIMODAL GRAND CHALLENGES

---===========
9-13 October 2023, Paris - France
https://icmi.acm.org/2023/

---===========

Teams are encouraged to submit proposals for one or more ICMI Multimodal Grand Challenges. The International Conference on Multimodal Interaction (ICMI) is the world's leading venue for multidisciplinary research on multimodal human-human and human-computer interaction, interfaces, and system development. Identifying the best algorithms and their failure modes is necessary for developing systems that can reliably interpret human-human communication or respond to human input. The availability of datasets and common goals has led to significant development in domains such as computer vision, speech recognition, computational (para-) linguistics, and physiological signal processing, for example. We invite the ICMI community to propose, define, and address the scientific Grand Challenges in our field during the next five years. The goal of the ICMI Multimodal Grand Challenges is to elicit fresh ideas from the ICMI community and to generate momentum for future collaborative efforts. Challenge tasks involving analysis, synthesis, and interaction are all feasible.

We invite organizers from various fields related to multimodal interaction to propose and run Grand Challenge events at ICMI 2023. We are looking for exciting and stimulating challenges including but not limited to the following categories:

  • Dataset-driven challenge.
    This challenge will provide a dataset that is exemplary of the complexities of current and future multimodal problems, and one or more multimodal tasks whose performance can be objectively measured and compared in rigorous conditions. Participants in the Challenge will evaluate their methods against the challenge data in order to identify areas of strengths and weaknesses.

  • System-driven challenge.
    This challenge will provide an interactive problem system (e.g. dialog-based or non-verbal-based) and the associated resources, which can allow people to participate through the integration of specific modules or alternative full systems. Proposers should also establish systematic evaluation procedures.

Prospective organizers should submit a five-page maximum proposal containing the following information:

  1. Title
  2. Abstract appropriate for possible Web promotion of the Challenge
  3. Distinctive topics to be addressed and specific goals
  4. Detailed description of the Challenge and its relevance to multimodal interaction
  5. Length (full day or half day)
  6. Plan for soliciting participation and list of potential participants
  7. Description of how submissions to the challenge will be evaluated, and a list of proposed reviewers
  8. Proposed schedule for releasing datasets (if applicable) and/or systems (if applicable) and receiving submissions.
  9. Short biography of the organizers (preferably from multiple institutions)
  10. Funding source (if any) that supports or could support the challenge organization
  11. Draft call for papers: affiliations and email address of the organizers; summary of the Grand Challenge; list of potential Technical Program Committee members and their affiliations, important dates

Proposals will be evaluated based on originality, ambition, feasibility, and implementation plan. A Challenge with dataset(s) or system(s) that has had pilot results to ensure its representativity and suitability to the proposed task will be given preference for acceptance; an additional 1 page description must be attached in such case. Continuation of or variants on previous ICMI grand challenges are welcome, though we ask for submissions of this form to highlight the number of participants that attended during the previous year and describe what changes (if any) will be made from the previous year.

The ICMI conference organizers will offer support with basic logistics, which includes rooms and equipment to run the challenge workshop, coffee breaks synchronized with the main track, etc.

Important Dates and Contact Details

---==

Proposals due: February 3, 2023
Proposal notification: February 10, 2023
Paper camera-ready: August 13, 2023
Grand challenge date: October 9 or 14, 2023

Proposals should be emailed to the ICMI 2023 Multimodal Grand Challenge Chairs, Sean Andrist and Fabien Ringeval:  icmi2023-grand-challenge-chairs@acm.orgmailto:icmi2023-grand-challenge-chairs@acm.org

Prospective organizers are also encouraged to contact the co-chairs if they have any questions.


Dr. Khiet Truong | Associate Professor | General Co-Chair Interspeech 2025
University of Twente | Human Media Interaction | Enschede, NL
Campus building Citadel, room H235 | T: +31 (0) 53 - 489 3683
http://khiettruong.space | https://people.utwente.nl/k.p.truong

ICMI'23 CALL FOR MULTIMODAL GRAND CHALLENGES ============================================ 9-13 October 2023, Paris - France https://icmi.acm.org/2023/ ============================================ Teams are encouraged to submit proposals for one or more ICMI Multimodal Grand Challenges. The International Conference on Multimodal Interaction (ICMI) is the world's leading venue for multidisciplinary research on multimodal human-human and human-computer interaction, interfaces, and system development. Identifying the best algorithms and their failure modes is necessary for developing systems that can reliably interpret human-human communication or respond to human input. The availability of datasets and common goals has led to significant development in domains such as computer vision, speech recognition, computational (para-) linguistics, and physiological signal processing, for example. We invite the ICMI community to propose, define, and address the scientific Grand Challenges in our field during the next five years. The goal of the ICMI Multimodal Grand Challenges is to elicit fresh ideas from the ICMI community and to generate momentum for future collaborative efforts. Challenge tasks involving analysis, synthesis, and interaction are all feasible. We invite organizers from various fields related to multimodal interaction to propose and run Grand Challenge events at ICMI 2023. We are looking for exciting and stimulating challenges including but not limited to the following categories: * Dataset-driven challenge. This challenge will provide a dataset that is exemplary of the complexities of current and future multimodal problems, and one or more multimodal tasks whose performance can be objectively measured and compared in rigorous conditions. Participants in the Challenge will evaluate their methods against the challenge data in order to identify areas of strengths and weaknesses. * System-driven challenge. This challenge will provide an interactive problem system (e.g. dialog-based or non-verbal-based) and the associated resources, which can allow people to participate through the integration of specific modules or alternative full systems. Proposers should also establish systematic evaluation procedures. Prospective organizers should submit a five-page maximum proposal containing the following information: 1. Title 2. Abstract appropriate for possible Web promotion of the Challenge 3. Distinctive topics to be addressed and specific goals 4. Detailed description of the Challenge and its relevance to multimodal interaction 5. Length (full day or half day) 6. Plan for soliciting participation and list of potential participants 7. Description of how submissions to the challenge will be evaluated, and a list of proposed reviewers 8. Proposed schedule for releasing datasets (if applicable) and/or systems (if applicable) and receiving submissions. 9. Short biography of the organizers (preferably from multiple institutions) 10. Funding source (if any) that supports or could support the challenge organization 11. Draft call for papers: affiliations and email address of the organizers; summary of the Grand Challenge; list of potential Technical Program Committee members and their affiliations, important dates Proposals will be evaluated based on originality, ambition, feasibility, and implementation plan. A Challenge with dataset(s) or system(s) that has had pilot results to ensure its representativity and suitability to the proposed task will be given preference for acceptance; an additional 1 page description must be attached in such case. Continuation of or variants on previous ICMI grand challenges are welcome, though we ask for submissions of this form to highlight the number of participants that attended during the previous year and describe what changes (if any) will be made from the previous year. The ICMI conference organizers will offer support with basic logistics, which includes rooms and equipment to run the challenge workshop, coffee breaks synchronized with the main track, etc. Important Dates and Contact Details =================================== Proposals due: February 3, 2023 Proposal notification: February 10, 2023 Paper camera-ready: August 13, 2023 Grand challenge date: October 9 or 14, 2023 Proposals should be emailed to the ICMI 2023 Multimodal Grand Challenge Chairs, Sean Andrist and Fabien Ringeval: icmi2023-grand-challenge-chairs@acm.org<mailto:icmi2023-grand-challenge-chairs@acm.org> Prospective organizers are also encouraged to contact the co-chairs if they have any questions. --- Dr. Khiet Truong | Associate Professor | General Co-Chair Interspeech 2025 University of Twente | Human Media Interaction | Enschede, NL Campus building Citadel, room H235 | T: +31 (0) 53 - 489 3683 http://khiettruong.space | https://people.utwente.nl/k.p.truong