MARCH: Modeling and Animating Realistic Crowds and Humans (2021)
Workshop in 4th IEEE International Conference on Artificial Intelligence and Virtual Reality (AIVR)
16 November 2021, Online (hosted in Taichung, Taiwan)
News
(2021/11/17) We've had a successful and interesting workshop yesterday. Thanks to all attendees, and we hope to see you next year!
(2021/11/15) The IEEE AIVR conference has started, and the MARCH workshop will take place tomorrow. We hope to see you there!
(2021/11/01) More details of the two keynote presentations are now available.
(2021/10/22) The full program of the workshop is now available. Please note that the workshop will take place fully online on Tuesday, 16 November 2021. Registration instructions can be found below. Authors will receive presentation instructions via e-mail soon.
(2021/10/01) Reviews and decisions have been sent to the authors last week. In line with the main IEEE AIVR conference, the submission deadline of camera-ready papers is now 20 October.
(2021/08/31) The submission deadline has been extended to 3 September.
(2021/08/23) We've added a link to the EasyChair paper submission page. Please see the section "Submission Instructions" below.
(2021/07/16) Our second keynote speaker will be Nuria Pelechano from Universitat Politecnica de Catalunya.
(2021/07/15) We're happy to announce that Jehee Lee from Seoul National University will be a keynote speaker at the workshop.
(2021/06/25) We are excited to host a new edition of the IEEE-MARCH workshop at AIVR 2021. More information will be added to this website soon. For reference, the website of the 2020 edition of MARCH is still available.
Introduction
Digital humans are key aspects of the rapidly evolving areas of virtual reality, augmented reality, virtual production, and gaming. They are becoming more and more commonly used for entertainment, education, health, security, and many other purposes. Modeling and animating digital humans are still challenging tasks that may lead to uncanny results if not properly achieved. Creating crowds and groups of digital humans adds another layer of complexity.
The objective of this Workshop on Modeling and Animating Realistic Crowds and Humans (MARCH) is to present the state of the art and new research perspectives on three key challenges: modeling digital humans, animating them, and creating groups/crowds of autonomous characters. The workshop aims to establish a new platform for the development of virtual-human technology at the intersection of Artificial intelligence (AI) and Virtual reality (VR). We expect that the workshop will provide an opportunity for researchers to develop new techniques, and will lead to new collaboration among the participants.
IEEE-MARCH will be a half-day workshop during the 4th IEEE International Conference on Artificial Intelligence and Virtual Reality (AIVR), which will be held on 15-17 November 2021 in Taichung, Taiwan. Due to the COVID-19 situation, the conference will be held largely online, and the workshop will be fully online.
Full Program
Date: Tuesday 16 November 2021.
All times are in UTC+8 (Taiwan time).
16.30-18.30 Part 1: Virtual Humans
16.30-16.40 Introduction
16.40-17.30 Keynote (Jehee Lee)
How to train your virtual dragon, human, and octopus via deep learning17.30-17.50 Paper presentation
Needs Model for an Autonomous Agent during Long-term Simulations
(Lysa Gramoli, Jérémy Lacoche, Anthony Foulonneau, Valérie Gouranton and Bruno Arnaldi)17.50-18.10 Paper presentation
A Multi-Agent Approach to Simulate Collaborative Classroom Activities using Digital Humans
(Wesley Deneke, Hannah Simurdak, Karl Truong, Andrew Strong and Andy Holland)18.10-18.30 Presentation (Funda Durupinar)
Expressive Virtual Humans
18.30-18.45 Break
18.45-20.45 Part 2: From Individuals to Crowds
18.45-19.35 Keynote (Nuria Pelechano)
Interacting with virtual humanoids: from self-avatars to crowds19.35-19.55 Paper presentation
Social Crowd Simulation: The Challenge of Fragmentation
(Michelangelo Diamanti and Hannes Högni Vilhjálmsson)19.55-20.15 Paper presentation
Representative Synthetic Crowds for Inclusive Environment Design
(Brandon Haworth, Mubbasir Kapadia and Petros Faloutsos)20.15-20.35 Presentation (Tomer Weiss)
Real-Time Virtual Content: From Spaces to Crowd Dynamics20.35-20.45 Closing remarks
Registration
The MARCH workshop will be an online video meeting during one day of the IEEE AIVR 2021 conference. If you register for the full conference, you get access to the MARCH workshop and to all other conference sessions. There is no separate registration procedure for the workshop itself.
Click here to see the registration instructions on the IEEE AIVR 2021 website.
Invited Speakers
Jehee Lee (Seoul National University, Republic of Korea)
Title: How to train your virtual dragon, human, and octopus via deep learning
Abstract: The human musculoskeletal system is made up of 206 bones and over 600 muscles. Even simple movements such as walking and jumping require careful coordination between hundreds of muscles, which includes sending the right amount of excitation signal to individual muscles to move the body forward while maintaining its balance. This poses the notoriously challenging problem of finding control policies for high-order, high-dimensional dynamical systems. The first half of the talk will discuss how to create a computer model of the human musculoskeletal system and control the simulation model to perform various movements leveraging deep reinforcement learning. Specifically, three technical components, predictive gait analysis, intelligent gait analysis, and patient-specific musculoskeletal modeling, will be discussed to deal with common problems encountered in computer graphics and computational biomechanics. The second half of the talk will be devoted to creating virtual humans and virtual animals via data-driven and physics-based learning frameworks. The topics include creating agile and time-critically responsive characters from a collection of unorganized human motion data (SIGGRAPH 2021), learning a family of motor skills from a single motion clip (SIGGRAPH 2021), and reconstructing 3D human motion from dynamic view video (SIGGRAPH Asia 2021). All the ideas and algorithms to animate virtual humans can also be used to create virtual creatures that fly in the air with flapping wings or swim underwater. I will briefly present the research results on virtual creatures that have been conducted in the SNU movement research lab over the past few years.
Bio: Jehee Lee is a full professor of Computer Science and Engineering at Seoul National University. His research interests span computer graphics, animation, computational biomechanics, machine learning and robotics. He is interested in developing new ways of understanding, representing, planning, and simulating human and animal movements. He co-chaired Pacific Graphics 2019, MIG 2018, and ACM/EG Symposium on Computer Animation in 2012. He has served as an associate editor of IEEE Transactions on Visualization and Computer Graphics and will be the program chair of SIGGRAPH Asia 2022. He heads the SNU Movement Research Laboratory.
Nuria Pelechano (Universitat Politecnica de Catalunya, Spain)
Title: Interacting with virtual humanoids: from self-avatars to crowds
Abstract: Virtual humans can be used in immersive VR fordifferent purposes, such as to represent the user with a self-avatar, to visualize other users for collaborative VR, or to simulate autonomous characters or crowds that serve the purpose of bringing life into the virtual environment. In order to make the virtual environment believable enough for the user to experience presence and behave as he/she would do in the real world, each category may have different requirements when it comes to the quality of the appearance, animations, simulation or behavioral response. For instance, self-avatars need animations that accurately track user movements, whereas avatars representing other users may simply need smooth animations even if they are not a perfect match of the other users’ posture. Crowd simulation may require high fidelity appearance and exhibit a variety of animations to make the environment believable. In this talk I will be presenting our recent work exploring what aspects of virtual humans are key to achieve populated virtual environments where users can behave in a natural manner, and how these environments can then be used to study human behavior or create a future for collaborative work and teaching.
Bio: Nuria Pelechano is an Associate Professor at the Universitat Politecnica de Catalunya, Spain, where she is a member of the Research Center for Visualization, Virtual Reality and Graphics Interaction. She is the co-author of two books on Crowd Simulation and has published in journals and conferences on Computer Graphics and Animation. She has participated in projects funded by the EU (currently being the research leader of the MSCA ITN-ETN CLIPE), Spanish Government, and USA institutions, and also worked on technology transfer projects regarding crowd evacuation, and applications of VR for architecture design. Her research interests include simulation, animation and rendering of crowds, generation of navigation meshes, and human-avatar interaction in Virtual Reality.
Scope
This workshop invites researchers to submit original, high-quality research papers related to modeling and animation of humans and/or crowds. The workshop welcomes both technical papers and position papers, and authors may also submit work in progress.
Relevant topics include, but are not limited to:
Modeling and animation of humanoid characters
Expressive character animation (e.g. face, gaze, gestures, ...)
ML and deep learning algorithms for character animation
Avatars in VR/AR
Social interaction and groups of virtual humans
Navigation for autonomous characters
Crowd simulation
Important Dates
Submission deadline:
27 August3 September 2021 (Anywhere on Earth)Notifications: 17 September 2021
Camera-ready deadline:
1 October20 October 2021Conference date: 15-17 November 2021
Workshop date: TBA
Submission Instructions
Authors are invited to submit a paper of up to 4 pages, in double-column IEEE format following the official IEEE Manuscript Formatting guidelines. Please anonymize your submissions, as the workshop uses a double-blind review process.
For each accepted paper, at least one author is expected to register for the IEEE AIVR conference and give a presentation at this workshop.
Accepted papers will be included in the conference proceedings and will appear on IEEE Xplore Digital Library. Also consider the IEEE policies for publications (i.e. you must own the copyright to all parts, and the manuscript must be original work and not currently under review elsewhere).
Papers can be submitted via the EasyChair page for IEEE AIVR. To reach this page, click on the button below:
When you start a new submission, you will first be asked to choose a track within the conference. Please make sure to choose the track "Workshop: Modeling and Animating Realistic Crowds and Humans".
Organizing Committee
Breda University of Applied Sciences, The Netherlands
Program Committee
Panayiotis Charalambous (CYENS - Centre of Excellence, Cyprus)
Ludovic Hoyet (Inria Rennes, France)
Anne-Hélène Olivier (Université Rennes 2, France)
Nuria Pelechano (Universitat Politecnica de Catalunya, Spain)
Jungdam Won (Facebook AI Research)
Contact
If you have any questions or remarks regarding this workshop, please contact Zerrin Yumak (Z.Yumak[at]uu.nl) or Wouter van Toll (wouter.van-toll[at]inria.fr).