A spatio-temporal 2D-models framework for human pose recovery in monocular sequences |
| |
| Authors: | Grégory Rogez Carlos Orrite-Uruñuela Jesús Martínez-del-Rincón |
| |
| Affiliation: | 1. Department of Electronics and Communication Engineering, Vardhaman College of Engineering (Autonomous), Hyderabad, India;2. Department of Electronics and Telecommunication, WCEM, Nagpur, India;3. Department of Electronics and Communication, Koneru Lakshmaiah Education Foundation, Hyderabad, India;1. NeurObesity Group, Department of Physiology, CIMUS, University of Santiago de Compostela-Instituto de Investigación Sanitaria, 15782 Santiago de Compostela, Spain;2. CIBER de Fisiopatología de la Obesidad y Nutrición (CIBERobn), 15706 Santiago de Compostela, Spain;1. Laboratory of Clinical Analysis and Biomechanics of Movement, University Hospital of Padova, Padova, Italy;2. NEUROMOVE-Rehab, Department of Neuroscience, University of Padova, Padova, Italy;3. MOVANT, Faculty of Medicine and Health Science, University of Antwerp, Antwerp, Belgium;4. Department of Paediatric Neuroscience, Fondazione IRCCS Istituto Neurologico Carlo Besta, Milano, Italy;5. Department of Pediatric Neurology, Antwerp University Hospital, Antwerp, Belgium;6. Pediatric Neurology, University Hospital of Verona, Verona, Italy;7. La Nostra Famiglia Association, University of Padova, Padova, Italy;8. Department of Information Engineering, University of Padova, Padova, Italy;9. Department of Statistical Sciences, University of Padova, Padova, Italy;10. Physical Medicine and Rehabilitation Unit, IRCCS - Istituto Ortopedico Rizzoli, Bologna, Italy;11. PNC, Padova Neuroscience Center, Padova, Italy |
| |
| Abstract: | This paper addresses the pose recovery problem of a particular articulated object: the human body. In this model-based approach, the 2D-shape is associated to the corresponding stick figure allowing the joint segmentation and pose recovery of the subject observed in the scene. The main disadvantage of 2D-models is their restriction to the viewpoint. To cope with this limitation, local spatio-temporal 2D-models corresponding to many views of the same sequences are trained, concatenated and sorted in a global framework. Temporal and spatial constraints are then considered to build the probabilistic transition matrix (PTM) that gives a frame to frame estimation of the most probable local models to use during the fitting procedure, thus limiting the feature space. This approach takes advantage of 3D information avoiding the use of a complex 3D human model. The experiments carried out on both indoor and outdoor sequences have demonstrated the ability of this approach to adequately segment pedestrians and estimate their poses independently of the direction of motion during the sequence. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
