|Topic||Human body-tracking for healthcare applications|
|Domain||Computer Vision (Computer Science, Artificial Intelligence)|
|Project||ECFvisuL: Évaluation automatique continuelle de la capacité fonctionnelle humaine à domicile à l’aide de la vision par ordinateur|
|Geographic location||Brest, France|
|Contract duration||18 months|
|Departement||RAMBO team, Computer Science Department, IMT Atlantique|
|Minimum degree required||Doctorate (PhD)|
|Starting date||As soon as possible, 1 January 2023 at latest|
|Gross annual salary||Between 34 400 € and 35 650 € (depending on your experience)|
|Application deadline||31 January 2022|
The postdoctorate will be located at IMT Atlantique, on its Brest campus, in the RAMBO team (member of Lab-STICC laboratory, mixed research unit of CNRS #6285) and supervised by the associate professors Mihai Andries and Panagiotis Papadakis. The research domains of the RAMBO team are Cognitive Robotics, Robot Learning, Ambient assisted living and Human-robot interaction. The team possesses a LivingLab for e-Health/Smart Living Experiment’HAAL, equipped with sensors for tracking human activity, where experiments related to the project can be performed, as well as a GERT ageing simulator.
This project is part of the RAMBO research activities on ageing-at-home and is complementary to our ongoing research on the recognition of activities of daily living (ADL).
Functional capacity evaluation (FCE) is a set of dynamical tests consisting of functional evaluations (sitting, kneeling, standing, bending, climbing a staircase, grasping an object, lifting, carrying, etc.) which provide information on the capacity of a person do act in a work environment. FCE is currently used by employers and medical professionals to decide when an employee is ready to return to work but it can be more broadly used as an indicator of human musculo-skeletal health.
We propose to develop an automatic solution for continuous FCE in a domestic environmenton the basis of a Computer Vision system for tracking human body posture. Such a system would acquire more temporal data on the musculo-skeletal capacity of a human than a typical doctorat during an FCE session, producing more robust evaluation results.
3. SKILLS AND ABILITIES:
- know statistics;
- know machine learning techniques for human-body posture recognition;
- know English Language (B2 level at least).
- know the scientific literature on musculo-skeletal disorders;
- know how to conceive experiments on physical capacity evaluation;
- know how to write scientific articles.
- apply machine learning techniques for human activity recognition;
- apply machine learning techniques for visual estimation of carried weight;
- present your research work to a scientific audience, as well as to the general public.
- use a 3D motion capture device (like Optitrack);
- use software for visual recognition of human-body posture (like MargiPose or Skeleton Tracking SDK);
- use inertial sensors.
- be able to work in a team;
- be able to communicate regularly with your colleagues and supervisors.
4. HOW TO SUBMIT YOUR APPLICATION ?
Please submit your application through the Recruitee system. You will need:
- your CV,
- a motivation letter,
- a copy of your doctorate diploma, and
- a list of referees (2-3).
Eligibility criteria apply regarding candidate’s mobility: the postdoctoral candidate must have spent at least 18 months outside of France between 1 May 2018 and the contract starting date.
Applications will be processed on a continuous basis.
The deadline for submitting your application is 31 January 2022.
- Devanne, Maxime, and Panagiotis Papadakis. "Recognition of activities of daily living via hierarchical long-short term memory networks." In 2019 IEEE International Conference on Systems, Man and Cybernetics (SMC), pp. 3318-3324. IEEE, 2019.
- C. Le Bono, P. Papadakis, C. Buche, Assessment of Conformal Use of Personal Protective Equipment by Object and Human Pose Recognition, IEEE International Conference on Safety, Security and Rescue Robotics (SSRR), 2020
- Andries, Mihai. "Object and human tracking, and robot control through a load sensing floor." PhD diss., Université de Lorraine, 2015.
Brest is a middle-sized town on the French Atlantic coast, home to a population of nearly 140 thousand people in 2018. Located near the western tip of the Brittany peninsula, it has a temperate oceanic climate, with mild summers and wet winters.
Almost entirely destroyed during WW2, the city center was rebuilt in Art Deco style. Historical monuments include the Brest castle, the Tanguy tower, and the Saint-Louis church. Gastronomically, Brittany is known for its galettes, crêpes, the Far breton and the Kouign Amann. Profiting from its beautiful coastline, Brest offers plenty of summer water activities. The Blancs Sablons beach, popular among the locals, is only 25 minutes away by car. The picturesque GR34 coastline trail is famous among hikers.
Brest hosts offices of Naval Group, THALES, Dassault Systemes, Ifremer, Université Bretagne occidentale and IMT Atlantique, among others. Brest is connected by high-speed rail to Rennes (2h) and Paris (3h45). The Brest international airport offers direct flights to Lisbon and Porto, Majorca, and Heraklion in Greece, as well as to national destinations like Nice, Lyon, and Toulouse.
As many of us, you may want to come for a year and ultimately stay for longer!