IEICE TRANSACTIONS on Information
Robust Visual Tracking via Coupled Randomness
Summary :
Tracking algorithms for arbitrary objects are widely researched in the field of computer vision. At the beginning, an initialized bounding box is given as the input. After that, the algorithms are required to track the objective in the later frames on-the-fly. Tracking-by-detection is one of the main research branches of online tracking. However, there still exist two issues in order to improve the performance. 1) The limited processing time requires the model to extract low-dimensional and discriminative features from the training samples. 2) The model is required to be able to balance both the prior and new objectives' appearance information in order to maintain the relocation ability and avoid the drifting problem. In this paper, we propose a real-time tracking algorithm called coupled randomness tracking (CRT) which focuses on dealing with these two issues. One randomness represents random projection, and the other randomness represents online random forests (ORFs). In CRT, the gray-scale feature is compressed by a sparse measurement matrix, and ORFs are used to train the sample sequence online. During the training procedure, we introduce a tree discarding strategy which helps the ORFs to adapt fast appearance changes caused by illumination, occlusion, etc. Our method can constantly adapt to the objective's latest appearance changes while keeping the prior appearance information. The experimental results show that our algorithm performs robustly with many publicly available benchmark videos and outperforms several state-of-the-art algorithms. Additionally, our algorithm can be easily utilized into a parallel program.
- Publication
- IEICE TRANSACTIONS on Information Vol.E98-D No.5 pp.1080-1088
- Publication Date
- 2015/05/01
- Publicized
- 2015/02/04
- Online ISSN
- 1745-1361
- DOI
- 10.1587/transinf.2014EDP7210
- Type of Manuscript
- PAPER
- Category
- Image Recognition, Computer Vision
Authors
Chao ZHANG
Iwate University
Yo YAMAGATA
Iwate University
Takuya AKASHI
Iwate University
Keyword
Latest Issue
Copyrights notice of machine-translated contents
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Cite this
Copy
Chao ZHANG, Yo YAMAGATA, Takuya AKASHI, "Robust Visual Tracking via Coupled Randomness" in IEICE TRANSACTIONS on Information,
vol. E98-D, no. 5, pp. 1080-1088, May 2015, doi: 10.1587/transinf.2014EDP7210.
Abstract: Tracking algorithms for arbitrary objects are widely researched in the field of computer vision. At the beginning, an initialized bounding box is given as the input. After that, the algorithms are required to track the objective in the later frames on-the-fly. Tracking-by-detection is one of the main research branches of online tracking. However, there still exist two issues in order to improve the performance. 1) The limited processing time requires the model to extract low-dimensional and discriminative features from the training samples. 2) The model is required to be able to balance both the prior and new objectives' appearance information in order to maintain the relocation ability and avoid the drifting problem. In this paper, we propose a real-time tracking algorithm called coupled randomness tracking (CRT) which focuses on dealing with these two issues. One randomness represents random projection, and the other randomness represents online random forests (ORFs). In CRT, the gray-scale feature is compressed by a sparse measurement matrix, and ORFs are used to train the sample sequence online. During the training procedure, we introduce a tree discarding strategy which helps the ORFs to adapt fast appearance changes caused by illumination, occlusion, etc. Our method can constantly adapt to the objective's latest appearance changes while keeping the prior appearance information. The experimental results show that our algorithm performs robustly with many publicly available benchmark videos and outperforms several state-of-the-art algorithms. Additionally, our algorithm can be easily utilized into a parallel program.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2014EDP7210/_p
Copy
@ARTICLE{e98-d_5_1080,
author={Chao ZHANG, Yo YAMAGATA, Takuya AKASHI, },
journal={IEICE TRANSACTIONS on Information},
title={Robust Visual Tracking via Coupled Randomness},
year={2015},
volume={E98-D},
number={5},
pages={1080-1088},
abstract={Tracking algorithms for arbitrary objects are widely researched in the field of computer vision. At the beginning, an initialized bounding box is given as the input. After that, the algorithms are required to track the objective in the later frames on-the-fly. Tracking-by-detection is one of the main research branches of online tracking. However, there still exist two issues in order to improve the performance. 1) The limited processing time requires the model to extract low-dimensional and discriminative features from the training samples. 2) The model is required to be able to balance both the prior and new objectives' appearance information in order to maintain the relocation ability and avoid the drifting problem. In this paper, we propose a real-time tracking algorithm called coupled randomness tracking (CRT) which focuses on dealing with these two issues. One randomness represents random projection, and the other randomness represents online random forests (ORFs). In CRT, the gray-scale feature is compressed by a sparse measurement matrix, and ORFs are used to train the sample sequence online. During the training procedure, we introduce a tree discarding strategy which helps the ORFs to adapt fast appearance changes caused by illumination, occlusion, etc. Our method can constantly adapt to the objective's latest appearance changes while keeping the prior appearance information. The experimental results show that our algorithm performs robustly with many publicly available benchmark videos and outperforms several state-of-the-art algorithms. Additionally, our algorithm can be easily utilized into a parallel program.},
keywords={},
doi={10.1587/transinf.2014EDP7210},
ISSN={1745-1361},
month={May},}
Copy
TY - JOUR
TI - Robust Visual Tracking via Coupled Randomness
T2 - IEICE TRANSACTIONS on Information
SP - 1080
EP - 1088
AU - Chao ZHANG
AU - Yo YAMAGATA
AU - Takuya AKASHI
PY - 2015
DO - 10.1587/transinf.2014EDP7210
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E98-D
IS - 5
JA - IEICE TRANSACTIONS on Information
Y1 - May 2015
AB - Tracking algorithms for arbitrary objects are widely researched in the field of computer vision. At the beginning, an initialized bounding box is given as the input. After that, the algorithms are required to track the objective in the later frames on-the-fly. Tracking-by-detection is one of the main research branches of online tracking. However, there still exist two issues in order to improve the performance. 1) The limited processing time requires the model to extract low-dimensional and discriminative features from the training samples. 2) The model is required to be able to balance both the prior and new objectives' appearance information in order to maintain the relocation ability and avoid the drifting problem. In this paper, we propose a real-time tracking algorithm called coupled randomness tracking (CRT) which focuses on dealing with these two issues. One randomness represents random projection, and the other randomness represents online random forests (ORFs). In CRT, the gray-scale feature is compressed by a sparse measurement matrix, and ORFs are used to train the sample sequence online. During the training procedure, we introduce a tree discarding strategy which helps the ORFs to adapt fast appearance changes caused by illumination, occlusion, etc. Our method can constantly adapt to the objective's latest appearance changes while keeping the prior appearance information. The experimental results show that our algorithm performs robustly with many publicly available benchmark videos and outperforms several state-of-the-art algorithms. Additionally, our algorithm can be easily utilized into a parallel program.
ER -
English
