Dictionary learning and confidence map estimation-based tracker for robot-assisted therapy system

Xiaolong Zhou; Sixian Chan; Junwei Li; Shengyong Chen; Honghai Liu

doi:10.1007/978-3-030-31654-9_13

Dictionary learning and confidence map estimation-based tracker for robot-assisted therapy system

Xiaolong Zhou^*
, Sixian Chan
, Junwei Li
, Shengyong Chen
, Honghai Liu

^*Corresponding author for this work

Quzhou University
Zhejiang University of Technology
Huawei Technologies Co., Ltd.
Tianjin University of Technology
University of Portsmouth

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

In this paper, we propose a new tracker based on dictionary learning and confidence map estimation for a robot-assisted therapy system. We first over-segment the image into superpixel patches, and then employ color and depth cues to estimate the object confidence of each superpixel patch. We build two Bag-of-Word (BoW) models from initial frames to encode foreground/background appearance, and compute object confidence at superpixel level using BoW model in both foreground and background. We further refine target confidence by depth-based statistical features to mitigate noise interference and the uncertainty of visual cues. We derive the global confidence of each target candidate at bag level, and incorporate the confidence estimations to determine the posterior probability of each candidate within the Bayesian framework. Experimental results demonstrate the superior performance of the proposed method, especially in long-term tracking and occlusion handling.

Original language	English
Title of host publication	Pattern Recognition and Computer Vision- 2nd Chinese Conference, PRCV 2019, Proceedings, Part I
Editors	Zhouchen Lin, Liang Wang, Tieniu Tan, Jian Yang, Guangming Shi, Nanning Zheng, Xilin Chen, Yanning Zhang
Publisher	Springer
Pages	147-159
Number of pages	13
ISBN (Print)	9783030316532
DOIs	https://doi.org/10.1007/978-3-030-31654-9_13
State	Published - 2019
Externally published	Yes
Event	2nd Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2019 - Xi'an, China Duration: 8 Nov 2019 → 11 Nov 2019

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	11857 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	2nd Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2019
Country/Territory	China
City	Xi'an
Period	8/11/19 → 11/11/19

Keywords

Bag-of-Word
Object tracking
Occlusion handling
RGB-D

Access to Document

10.1007/978-3-030-31654-9_13

Cite this

Zhou, X., Chan, S., Li, J., Chen, S., & Liu, H. (2019). Dictionary learning and confidence map estimation-based tracker for robot-assisted therapy system. In Z. Lin, L. Wang, T. Tan, J. Yang, G. Shi, N. Zheng, X. Chen, & Y. Zhang (Eds.), Pattern Recognition and Computer Vision- 2nd Chinese Conference, PRCV 2019, Proceedings, Part I (pp. 147-159). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11857 LNCS). Springer. https://doi.org/10.1007/978-3-030-31654-9_13

Zhou, Xiaolong ; Chan, Sixian ; Li, Junwei et al. / Dictionary learning and confidence map estimation-based tracker for robot-assisted therapy system. Pattern Recognition and Computer Vision- 2nd Chinese Conference, PRCV 2019, Proceedings, Part I. editor / Zhouchen Lin ; Liang Wang ; Tieniu Tan ; Jian Yang ; Guangming Shi ; Nanning Zheng ; Xilin Chen ; Yanning Zhang. Springer, 2019. pp. 147-159 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{b3c9e2eda41f40f8b365fede63225a0b,

title = "Dictionary learning and confidence map estimation-based tracker for robot-assisted therapy system",

abstract = "In this paper, we propose a new tracker based on dictionary learning and confidence map estimation for a robot-assisted therapy system. We first over-segment the image into superpixel patches, and then employ color and depth cues to estimate the object confidence of each superpixel patch. We build two Bag-of-Word (BoW) models from initial frames to encode foreground/background appearance, and compute object confidence at superpixel level using BoW model in both foreground and background. We further refine target confidence by depth-based statistical features to mitigate noise interference and the uncertainty of visual cues. We derive the global confidence of each target candidate at bag level, and incorporate the confidence estimations to determine the posterior probability of each candidate within the Bayesian framework. Experimental results demonstrate the superior performance of the proposed method, especially in long-term tracking and occlusion handling.",

keywords = "Bag-of-Word, Object tracking, Occlusion handling, RGB-D",

author = "Xiaolong Zhou and Sixian Chan and Junwei Li and Shengyong Chen and Honghai Liu",

note = "Publisher Copyright: {\textcopyright} Springer Nature Switzerland AG 2019.; 2nd Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2019 ; Conference date: 08-11-2019 Through 11-11-2019",

year = "2019",

doi = "10.1007/978-3-030-31654-9\_13",

language = "英语",

isbn = "9783030316532",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer",

pages = "147--159",

editor = "Zhouchen Lin and Liang Wang and Tieniu Tan and Jian Yang and Guangming Shi and Nanning Zheng and Xilin Chen and Yanning Zhang",

booktitle = "Pattern Recognition and Computer Vision- 2nd Chinese Conference, PRCV 2019, Proceedings, Part I",

address = "德国",

}

Zhou, X, Chan, S, Li, J, Chen, S & Liu, H 2019, Dictionary learning and confidence map estimation-based tracker for robot-assisted therapy system. in Z Lin, L Wang, T Tan, J Yang, G Shi, N Zheng, X Chen & Y Zhang (eds), Pattern Recognition and Computer Vision- 2nd Chinese Conference, PRCV 2019, Proceedings, Part I. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11857 LNCS, Springer, pp. 147-159, 2nd Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2019, Xi'an, China, 8/11/19. https://doi.org/10.1007/978-3-030-31654-9_13

Dictionary learning and confidence map estimation-based tracker for robot-assisted therapy system. / Zhou, Xiaolong; Chan, Sixian; Li, Junwei et al.
Pattern Recognition and Computer Vision- 2nd Chinese Conference, PRCV 2019, Proceedings, Part I. ed. / Zhouchen Lin; Liang Wang; Tieniu Tan; Jian Yang; Guangming Shi; Nanning Zheng; Xilin Chen; Yanning Zhang. Springer, 2019. p. 147-159 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11857 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Dictionary learning and confidence map estimation-based tracker for robot-assisted therapy system

AU - Zhou, Xiaolong

AU - Chan, Sixian

AU - Li, Junwei

AU - Chen, Shengyong

AU - Liu, Honghai

N1 - Publisher Copyright: © Springer Nature Switzerland AG 2019.

PY - 2019

Y1 - 2019

N2 - In this paper, we propose a new tracker based on dictionary learning and confidence map estimation for a robot-assisted therapy system. We first over-segment the image into superpixel patches, and then employ color and depth cues to estimate the object confidence of each superpixel patch. We build two Bag-of-Word (BoW) models from initial frames to encode foreground/background appearance, and compute object confidence at superpixel level using BoW model in both foreground and background. We further refine target confidence by depth-based statistical features to mitigate noise interference and the uncertainty of visual cues. We derive the global confidence of each target candidate at bag level, and incorporate the confidence estimations to determine the posterior probability of each candidate within the Bayesian framework. Experimental results demonstrate the superior performance of the proposed method, especially in long-term tracking and occlusion handling.

AB - In this paper, we propose a new tracker based on dictionary learning and confidence map estimation for a robot-assisted therapy system. We first over-segment the image into superpixel patches, and then employ color and depth cues to estimate the object confidence of each superpixel patch. We build two Bag-of-Word (BoW) models from initial frames to encode foreground/background appearance, and compute object confidence at superpixel level using BoW model in both foreground and background. We further refine target confidence by depth-based statistical features to mitigate noise interference and the uncertainty of visual cues. We derive the global confidence of each target candidate at bag level, and incorporate the confidence estimations to determine the posterior probability of each candidate within the Bayesian framework. Experimental results demonstrate the superior performance of the proposed method, especially in long-term tracking and occlusion handling.

KW - Bag-of-Word

KW - Object tracking

KW - Occlusion handling

KW - RGB-D

UR - https://www.scopus.com/pages/publications/85086143340

U2 - 10.1007/978-3-030-31654-9_13

DO - 10.1007/978-3-030-31654-9_13

M3 - 会议稿件

AN - SCOPUS:85086143340

SN - 9783030316532

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 147

EP - 159

BT - Pattern Recognition and Computer Vision- 2nd Chinese Conference, PRCV 2019, Proceedings, Part I

A2 - Lin, Zhouchen

A2 - Wang, Liang

A2 - Tan, Tieniu

A2 - Yang, Jian

A2 - Shi, Guangming

A2 - Zheng, Nanning

A2 - Chen, Xilin

A2 - Zhang, Yanning

PB - Springer

T2 - 2nd Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2019

Y2 - 8 November 2019 through 11 November 2019

ER -

Zhou X, Chan S, Li J, Chen S, Liu H. Dictionary learning and confidence map estimation-based tracker for robot-assisted therapy system. In Lin Z, Wang L, Tan T, Yang J, Shi G, Zheng N, Chen X, Zhang Y, editors, Pattern Recognition and Computer Vision- 2nd Chinese Conference, PRCV 2019, Proceedings, Part I. Springer. 2019. p. 147-159. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-31654-9_13

Dictionary learning and confidence map estimation-based tracker for robot-assisted therapy system

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this