By continuing to browse the site, you are agreeing to our use of cookies. (More details)

Phd - Analysis and optimization of elite athletes motor skills using markerless motion capture M/F

This offer is available in the following languages:
Français - Anglais

Ensure that your candidate profile is correct before applying. Your profile information will be added to the details for each application. In order to increase your visibility on our Careers Portal and allow employers to see your candidate profile, you can upload your CV to our CV library in one click!

Faites connaître cette offre !

General information

Reference : UPR3346-NADMAA-012
Workplace : FUTUROSCOPE
Date of publication : Tuesday, August 13, 2019
Scientific Responsible name : M. DOMALAIN - L. REVERET
Type of Contract : PhD Student contract / Thesis offer
Contract Period : 36 months
Start date of the thesis : 1 November 2019
Proportion of work : Full time
Remuneration : 2 135,00 € gross monthly

Description of the thesis topic

CNRS, at the Pprime laboratory (CNRS, UP, ISAE-ENSMA) in Poitiers-Futuroscope, recruits a PhD student as part of a research program that brings together the Institut PPRIME (UPR3346, Poitiers), the LJK (UMR. 5224,Grenoble), the Fédération Française de Cyclisme et the Grand Insep network (CREPS de Poitiers and TSF Voiron).The overall objective of this research program is to optimize performance in BMX Race.

The specific objective of this PhD is to help equip coaches and athletes with a methodological background that canbe used more independently for training (or even in competition). This methodological background includes the development of technological solutions in terms of "no marker" motion capture systems and data analysis algorithms. These algorithms will be based on both biomechanical models and image processing. Several sub-objectives were identified.

1) Study of the sensitivity of (bio)mechanical determinants of performance to the quality of experimental input data: the objective here is to assess the sensitivity of the output parameters of biomechanical modeling to experimental input data and model parameters in order to assess the extent to which the data acquisition protocol can be simplified.

2) Development of a biomechanical model adapted to degraded input data: the PhD will develop a biomechanical model that is more suited to degraded input data such as those from a video capture of movement without markers.To do this, new methods of calculations from mechanical and robotic approaches (e.g. Use of kinematic chains withloop closure constraints) and statistical analysis of the database will allow the development of a more robust model to the vagueness of experimental data.

3) Testing, improving and adapting existing non marker motion capture solutions: the most recent techniques foranalyzing without markers are based on machine learning.
On the basis of the work of the Jean Kuntzmann laboratory in this field, the PhD will choose the most suitable technological solution (camera system, sensors/complementary measurement tools) and adapt the existing algorithms in terms of image analysis to the specific needs identified in previous objectives.

The PhD student will benefit from a supportive material environment (already acquired and operational measurement tools, access to the structures of VSF and FFC) as well as experience gained during the first 4 years of the project (a thesis already in progress).

Work Context

The PhD is part of a collaborative project that brings together two research teams :

Pprime Institute (P') is a research unit affiliated to CNRS (National Center for Scientific Research) and the University of Poitiers. It is composed of more than 600 people whose research areas relate to Engineering Sciences and Materials Physics. The Robioss Team conducts research on multibody system dynamics and human motor performance. It has worked with athletes and sport federations for 20 years.

The LJK is a research unit affiliated to CNRS, UGA, Grenoble-INP and INRIA (National Institute of Research in Computer Science and Automation). The laboratory covers several areas in digital modeling and image processing. Within the LJK, Lionel Reveret is an expert in motion analysis through experimental approaches, real-time algorithms, 3D graphical visualization and user interface. He accompanies the BMX development projects at TSF-Voiron for his new BMX sport-study section and his new training track project

The doctoral student will be supervised by Lionel Reveret (thesis supervisor, LJK) and Mathieu Domalain (co-director, Pprime). He will be attached to the doctoral school Mathematics, Sciences and Technologies of Information, Computer Science (Grenoble).

The Phd student must have a Master's degree or an engineering degree with a specialization in motion analysis (biomechanics, video signal processing, applied mathematics) and must possess the following skills:
• MS degree (or engineer diploma) that includes: computer science, information technology, biomechanics, physics-based animation or a related discipline
• Demonstrated ability to conduct analysis of experimental data and perform mechanical simulations using programming
• Knowledge of biomechanical analysis of human motion (inverse kinematics, forward dynamics, etc.)
• Demonstrated programming skills (Python, C++ ; Matlab)
• Demonstrated ability to write technical/scientific reports in English
• Global interest in cycling and/or elite athlete's performance
• Fluent in French and/or English (basic knowledge of French is a plus)

Constraints and risks

The PhD candidate will spend some periods of time in Grenoble and in Poitiers (number and duration to be discussed and as a function of candidate background and personal convenience).
Travel will be planned in France and abroad.

We talk about it on Twitter!