Human motion tracking by combining view-based and model-based methods for monocular video sequences
- Authors
- Park, J.; Park, S.; Aggarwal, J.K.
- Issue Date
- 2003
- Publisher
- Springer Verlag
- Citation
- Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), v.2669, pp.650 - 659
- Journal Title
- Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
- Volume
- 2669
- Start Page
- 650
- End Page
- 659
- URI
- https://scholarworks.bwise.kr/hongik/handle/2020.sw.hongik/26608
- DOI
- 10.1007/3-540-44842-x_66
- ISSN
- 0302-9743
- Abstract
- Reliable tracking of moving humans is essential to motion estimation, video surveillance and human-computer interface. This paper presents a new approach to human motion tracking that combines view-based and model-based techniques. Monocular color video is processed at both pixel level and object level. At the pixel level, a Gaussian mixture model is used to train and classify individual pixel colors. At the object level, a 3D human body model projected on a 2D image plane is used to fit the image data. Our method does not use inverse kinematics due to the singularity problem. While many others use stochastic sampling for model-based motion tracking, our method is purely dependent on parameter optimization. We convert the human motion tracking problem into a parameter optimization problem. A cost function for parameter optimization is used to estimate the degree of the overlapping between the foreground input image silhouette and a projected 3D model body silhouette. The overlapping is computed using computational geometry by converting a set of pixels from the image domain to a polygon in the real projection plane domain. Our method is used to recognize various human motions. Motion tracking results from video sequences are very encouraging. © Springer-Verlag 2003.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - ETC > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/hongik/handle/2020.sw.hongik/26608)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.