Markerless computer methods for biomechanic analysis of human movement

• Grant number: 12/15811-1
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: April 01, 2013
• End date: May 31, 2014
• Field of knowledge: Health Sciences - Physical Education
• Principal Investigator: Ricardo Machado Leite de Barros
• Grantee: Fábio Augusto Salve Dias
• Host Institution: Faculdade de Educação Física (FEF). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil

Abstract

In this project we approach the problem of markerless biomecanic analysis of human movement, considering specifically as applications the gait analysis in a rehabilitation context and underwater situations, such as swimming. By using markerless methods, we hope to increase the possible applications of the analysis, but the level of accuracy needed for significant results increases the complexity of the problem. To this end, we divided the problem, and our proposal, in three parts: camera calibration, tridimensional reconstruction and pose recovery.The considered camera calibration has direct impact on the quality of the final result. Our proposal includes a new method for distortion modelling, which we hope increases the accuracy of the method and also includes methods to allow the use of moving cameras. The combined use of moving and static cameras is also a proposed development for the tridimensional reconstruction, along with improvements on the detection of moving objects. For the pose recovery, we plan to consider the quadridimensional space composed of the tridimensional reconstruction over time, in which we will fit a moving human body model, also quadridimensional. We hope such approach will reduce the positioning error of the articulations by enforcing the smoothness of the movement. To the best of our knowledge, such proposals are original.

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

DIAS, F.; COUSTY, J.; NAJMAN, L.. Dimensional operators for mathematical morphology on simplicial complexes. PATTERN RECOGNITION LETTERS, v. 47, p. 111-119, OCT 1 2014. (12/15811-1)