Model-Based Reinforcement of Kinect Depth Data for Human Motion Capture Applications
Motion capture systems have recently experienced a strong evolution. New cheap depth sensors and open source frameworks, such as OpenNI, allow for perceiving human motion on-line without using invasive systems. However, these proposals do not evaluate the validity of the obtained poses. This paper a...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2013-07-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1424-8220/13/7/8835 |
| _version_ | 1828109395285770240 |
|---|---|
| author | Andreas Skiadopoulos Pablo Bustos Juan Pedro Bandera Luis Vicente Calderita |
| author_facet | Andreas Skiadopoulos Pablo Bustos Juan Pedro Bandera Luis Vicente Calderita |
| author_sort | Andreas Skiadopoulos |
| collection | DOAJ |
| description | Motion capture systems have recently experienced a strong evolution. New cheap depth sensors and open source frameworks, such as OpenNI, allow for perceiving human motion on-line without using invasive systems. However, these proposals do not evaluate the validity of the obtained poses. This paper addresses this issue using a model-based pose generator to complement the OpenNI human tracker. The proposed system enforces kinematics constraints, eliminates odd poses and filters sensor noise, while learning the real dimensions of the performer’s body. The system is composed by a PrimeSense sensor, an OpenNI tracker and a kinematics-based filter and has been extensively tested. Experiments show that the proposed system improves pure OpenNI results at a very low computational cost. |
| first_indexed | 2024-04-11T11:01:40Z |
| format | Article |
| id | doaj.art-27b53d0100b14955bb7fb422bb39b6aa |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-11T11:01:40Z |
| publishDate | 2013-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-27b53d0100b14955bb7fb422bb39b6aa2022-12-22T04:28:30ZengMDPI AGSensors1424-82202013-07-011378835885510.3390/s130708835Model-Based Reinforcement of Kinect Depth Data for Human Motion Capture ApplicationsAndreas SkiadopoulosPablo BustosJuan Pedro BanderaLuis Vicente CalderitaMotion capture systems have recently experienced a strong evolution. New cheap depth sensors and open source frameworks, such as OpenNI, allow for perceiving human motion on-line without using invasive systems. However, these proposals do not evaluate the validity of the obtained poses. This paper addresses this issue using a model-based pose generator to complement the OpenNI human tracker. The proposed system enforces kinematics constraints, eliminates odd poses and filters sensor noise, while learning the real dimensions of the performer’s body. The system is composed by a PrimeSense sensor, an OpenNI tracker and a kinematics-based filter and has been extensively tested. Experiments show that the proposed system improves pure OpenNI results at a very low computational cost.http://www.mdpi.com/1424-8220/13/7/8835human motion capturesensorRGB-D sensorsrange camerapose analysis |
| spellingShingle | Andreas Skiadopoulos Pablo Bustos Juan Pedro Bandera Luis Vicente Calderita Model-Based Reinforcement of Kinect Depth Data for Human Motion Capture Applications Sensors human motion capture sensor RGB-D sensors range camera pose analysis |
| title | Model-Based Reinforcement of Kinect Depth Data for Human Motion Capture Applications |
| title_full | Model-Based Reinforcement of Kinect Depth Data for Human Motion Capture Applications |
| title_fullStr | Model-Based Reinforcement of Kinect Depth Data for Human Motion Capture Applications |
| title_full_unstemmed | Model-Based Reinforcement of Kinect Depth Data for Human Motion Capture Applications |
| title_short | Model-Based Reinforcement of Kinect Depth Data for Human Motion Capture Applications |
| title_sort | model based reinforcement of kinect depth data for human motion capture applications |
| topic | human motion capture sensor RGB-D sensors range camera pose analysis |
| url | http://www.mdpi.com/1424-8220/13/7/8835 |
| work_keys_str_mv | AT andreasskiadopoulos modelbasedreinforcementofkinectdepthdataforhumanmotioncaptureapplications AT pablobustos modelbasedreinforcementofkinectdepthdataforhumanmotioncaptureapplications AT juanpedrobandera modelbasedreinforcementofkinectdepthdataforhumanmotioncaptureapplications AT luisvicentecalderita modelbasedreinforcementofkinectdepthdataforhumanmotioncaptureapplications |