Research Internship Human Motion Representation Based on Labanotation H/F INRIA

Research Internship Human motion representation based on Labanotation
Le descriptif de l'offre ci-dessous est en Anglais
Type de contrat : Convention de stage

Niveau de diplôme exigé : Bac +4 ou équivalent

Fonction : Stagiaire de la recherche

A propos du centre ou de la direction fonctionnelle

The Inria Grenoble research center groups together almost 600 people in 27 research teams and 8 research support departments.

Staff is present on three campuses in Grenoble, in close collaboration with other research and higher education institutions (University Grenoble Alpes, CNRS, CEA, INRAE, ...), but also with key economic players in the area.

Inria Grenoble is active in the fields of high-performance computing, verification and embedded systems, modeling of the environment at multiple levels, and data science and artificial intelligence. The center is a top-level scientific institute with an extensive network of international collaborations in Europe and the rest of the world.

Contexte et atouts du poste

The internship will take place within the Morpheo research team at Inria of Grenoble Alpes University.

The position will start in early 2026 and be a 6 month M2 internship co-supervised by Sergi Pujades and Stefanie Wuhrer.

Mission confiée

Context:

It is now possible to acquire dynamic human motion sequences with high accuracy using platforms such as Kinovis [1] at Inria Grenoble, and large-scale datasets of different humans performing different motions have been captured and shared with the research community [2]. However, the question how to represent human motion continues to be studied. Existing works often proceed by using parametric human models of static poses, and by building motion models on top of this representation, e.g. [3,4]. Zhu et al. [5] provide a recent survey on this topic.

The goal of this internship is to study whether the use of dance notation can provide a suitable representation of human motion. In particular, we will study the use of Labanotation for this purpose. Labanotation provides a representation of human motion, mostly used in modern dance, that is localized in space and time. This representation is independent from morphology in the sense that (1) an expert can notate motions from an arbitrary dancer in Labanotation and (2) any dancer familiar with Labanotation can reproduce a specific dance based on a Labanotation score. It has already been successfully used to study the perception of human motion similarity [6] and in robotics applications [7].

Objectives and working directions:

The two main objectives of this internship are:

- Implement and test a system that given as input a 4D motion sequence (i.e. geometry of minimally dressed character), can automatically and robustly compute its representation in Laban notation.
- Implement and test a system that given a representation in Laban notation along with a rigged character, can automatically animate the character to perform the encoded motion.

Regular meetings with the supervisors will be organized.

References:
- https://kinovis.inria.fr/
- Armando, L. Boissieux, E. Boyer, J.-S. Franco, M. Humenberger, C. Legras, V. Leroy, M. Marsot, J. Pansiot, S. Pujades, R. Rekik Dit Nekhili, G. Rogez, A. Swamy, S. Wuhrer. 4DHumanOutfit: a multi-subject 4D dataset of human motion sequences in varying outfits exhibiting large displacements. Computer Vision and Image Understanding, 2023.
- Rempe, T. Birdal, A. Hertzmann, J. Yang, S. Sridhar, L. Guibas. HuMoR: 3D Human Motion Model for Robust Pose Estimation. International Conference on Computer Vision, 2021.
- Marsot, S. Wuhrer, J.-S. Franco, S. Durocher. A Structured Latent Space for Human Body Motion Generation. International Conference on 3D Vision, 2022.
- Zhu, X. Ma, D. Ro, H. Ci, J. Zhang, J. Shi, F. Gao, Q. Tian, Y. Wang. Human Motion Generation: A Survey. Transactions on Pattern Analysis and Machine Intelligence, 2024.
- Durupinar. Perception of human motion similarity based on laban movement analysis. ACM Symposium on Applied Perception, 2021.
- Ikeuchi, K., Ma, Z., Yan, Z., Kodoh, S., Nakamura, M. Describing Upper-Body Motions Based on Labanotation for Learning-from-Observation Robots. International Journal of Computer Vision 126, 1415-1429 (2018).

Principales activités

- Reading and analyzing advantages and disadvantages of related literature
- Studying and preparing data
- Implementing solutions for the problem
- Evaluating the implemented solutions and existing works

Compétences

The following skills and interests will be valued:

- Python programming
- Mathematical modeling
- Experience with deep learning is a plus
- Strong motivation
- Good collaboration capabilities (international context)

Avantages

- Subsidized meals
- Partial reimbursement of public transport costs
- Possibility of teleworking
- Professional equipment available (videoconferencing, loan of computer equipment, etc.)
- Social, cultural and sports events and activities

Rémunération

Minimum legal gratification