GENEA Challenge
Title
The GENEA Challenge 2023: Full-body speech-driven gesture generation in a dyadic setting
Website
https://genea-workshop.github.io/2023/challenge/
Abstract
The state-of-the-art in co-speech gesture generation is difficult to assess, since every research group tends to use their own data, embodiment, and evaluation methodology. To better understand and compare methods for gesture generation and evaluation, we are continuing the GENEA (Generation and Evaluation of Non-verbal Behaviour for Embodied Agents) Challenge, wherein different gesture-generation approaches are evaluated side by side in a large user study. This 2023 challenge will be hosted as a Multimodal Grand Challenge at ICMI 2023 and is a follow-up to the first and second editions of the GENEA Challenge, arranged in 2020 and 2022.
This year the challenge will focus on gesture synthesis in a dyadic setting, i.e., gestures that depend not only on speech, but also on the behaviour of an interlocutor in a conversation. We invite researchers in academia and industry working on any form of corpus-based non-verbal behaviour generation and gesticulation to submit entries to the challenge, whether their method is driven by rules or machine learning. Participants are provided a large, common dataset of speech and 3D motion to develop their systems, and then use these systems to generate motion on given test inputs. The generated motion clips are rendered onto a common virtual agent and evaluated for motion quality and appropriateness (with respect to the associated speech and to the interlocutor behaviour) in a crowdsourced user study.
Tentative timeline
May 1 – challenge training dataset released to participants
June 7 – test input released to participants
June 14 – deadline for participants to submit generated motion, start of evaluation
July 3 – release of crowdsourced evaluation results to participants
July 14 – paper submission deadline
August 4 – paper notification
August 11 – camera-ready version
October 9–13 – challenge presentations at ICMI
Contact
genea-challenge@googlegroups.com
Organizers
Dr. Taras Kucherenko, SEED – Electronic Arts, Stockholm, Sweden
Dr. Youngwoo Yoon, ETRI, South Korea
Rajmund Nagy, KTH Royal Institute of Technology, Stockholm, Sweden
Jieyeon Woo, Sorbonne University, France
Dr. Gustav Eje Henter, KTH Royal Institute of Technology, Stockholm, Sweden
EmotiW Challenge
Title
Emotion Recognition in the Wild Challenge (EmotiW23)
Website
https://sites.google.com/view/emotiw2023
Abstract
The Ninth Emotion Recognition in the Wild 2023 Grand Challenge consists of an all-day event with a focus on affective sensing in unconstrained conditions. This year EmotiW will host two sub-challenges: audio-visual modalities based group-level emotion classification sub-challenge and user engagement prediction sub-challenge. Traditionally, emotion recognition has been performed on laboratory controlled data. While undoubtedly worthwhile at the time, such lab controlled data poorly represents the environment and conditions faced in real-world situations. With the increase in the number of online shared media, it is worthwhile to explore the performance of emotion recognition methods that work ‘in the wild’.
Tentative timeline
Train and validate data available – March 2023
Test data available – June 2023
Last date for uploading the results – June 2023
Paper submission deadline – July 2023
Paper notification – 30 July 2023
Camera-ready papers – to be determined
All deadlines follow 11.59 PM PMT time.
Organizers
Abhinav Dhall, Indian Institute of Technology, Ropar and Monash University
Roland Goecke, University of Canberra
Tom Gedeon, Curtin University
Donghuo Zeng, KDDI Research
Yanan Wang, KDDI Research
Kazushi Ikeda, KDDI Research