Cell-Based Object Tracking Method for 3D Shape Reconstruction Using Multi-Viewpoint Active Cameras
Abstract
3D shape of objects can provide richer information for detecting, tracking or identifying the objects than a single 2D image of them. We tackle the 3D shape and texture reconstruction of an object moving in a widespread space using multi-viewpoint active cameras. Considering 3D shape and texture reconstruction, the problem in existing tracking methods using active cameras is that they cannot calibrate the active cameras accurately. We propose a cell-based tracking method that can produce multi-viewpoint images and accurate camera parameters for every frame. Our idea is to divide the space into cells and perform active camera control and calibration based on the cells. We demonstrate the performance of our method by simulation.
Cite
Text
Yamaguchi et al. "Cell-Based Object Tracking Method for 3D Shape Reconstruction Using Multi-Viewpoint Active Cameras." IEEE/CVF International Conference on Computer Vision Workshops, 2009. doi:10.1109/ICCVW.2009.5457460Markdown
[Yamaguchi et al. "Cell-Based Object Tracking Method for 3D Shape Reconstruction Using Multi-Viewpoint Active Cameras." IEEE/CVF International Conference on Computer Vision Workshops, 2009.](https://mlanthology.org/iccvw/2009/yamaguchi2009iccvw-cellbased/) doi:10.1109/ICCVW.2009.5457460BibTeX
@inproceedings{yamaguchi2009iccvw-cellbased,
title = {{Cell-Based Object Tracking Method for 3D Shape Reconstruction Using Multi-Viewpoint Active Cameras}},
author = {Yamaguchi, Tatsuhisa and Nobuhara, Shohei and Matsuyama, Takashi},
booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
year = {2009},
pages = {1306-1313},
doi = {10.1109/ICCVW.2009.5457460},
url = {https://mlanthology.org/iccvw/2009/yamaguchi2009iccvw-cellbased/}
}