Gait recognition is particularly studied as it may enable identification at distance.
La reconnaissance de la démarche est particulièrement étudiée car elle pourrait permettre l'identification de personne à distance.
The final objective is to recognize persons using standard cameras in any conditions. Most of the studies are still at laboratory level, capturing gait samples under controlled conditions.
L'objectif final est de reconnaitre les gens en utilisant de simple caméras dans n'importe quelles conditions. La plupart des études sont encore au niveau du laboratoire, et commencent par enregistrer des vidéos de démarche dans des conditions bien controlées.
- (1998) Recognizing People by Their Gait: The Shape of Motion / James J. Little, Jeffrey E. Boyd
- (2018) Biometric recognition by gait: A survey of modalities and features / Patrick Connor, Arun Ross.
Algorithms for gait recognition are at development stage, here are some links for more information:
Les algorithmes de reconnaissance de la démarche sont en développement, voici quelques liens pour en savoir plus:
- Univ of southampton, ISIS group (UK)
- Georgia Institute of Technology (US)
- [http://www-mitpress.mit.edu/e-journals/Videre/001/articles/Little-Boyd/gait/gait.html] MIT Recognizing people by their gait (with source code)
- Indian Institute of Technology Kanpur
- Video based gait analysis (Ronkonnen) Lappeenranta University of Technology
- Radar
Not only video cameras may be used to capture gait:
Les caméras vidéo ne sont pas les seules utilisables pour enregistrer la démarche:
Georgia Tech Research Institute researchers monitored 70 subjects as they walked past a laboratory-based radar system similar to the radar guns carried by police officers.
L'institut de recherche technologique de la Géorgie (US) enregistre 70 sujets marchant devant un radar de laboratoire similaire à ceux utilisés par la police.
- Accelerometers:
(2005) Identifying Users of Portable Devices from Gait Pattern with Accelerometers Mäntyjärvi & als (VTT)
(2018) I Walk, Therefore I Am: Continuous User Authentication with Plantar Biometrics Kuo-Hui Yeh ; Chunhua Su ; Wayne Chiu ; Lu Zhou
- Sensor enhanced floor
Towards the adaptive identification of walkers Suutala, Röning (Oulu Univ.) - Identascan
[http://www.identascan.com/] Identascan (Dr Li Liu / McGill Univ.) Develop human identification technologies for surveillance and security systems, which combine advantages of biometrics (face recognition, gait recognition) and human motion analysis (motion analysis of human body parts, human motion tracking) to recognize a person in a crowd or singularly.
Monty Pythons, sure :-)