The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Baru-baru ini, prestasi penjejak penapis korelasi diskriminatif (CF) semakin baik dan lebih baik dalam penjejakan visual. Dalam kertas kerja ini, kami mencadangkan penyelarasan spatial-temporal dengan anggaran keadaan yang tepat berdasarkan penapis korelasi diskriminatif (STPSE) untuk mencapai prestasi penjejakan yang lebih ketara. Pertama, kami mempertimbangkan perubahan berterusan keadaan objek, menggunakan maklumat daripada dua penapis sebelumnya untuk melatih model penapis korelasi. Di sini, kami melatih model penapis korelasi dengan ciri buatan tangan. Kedua, kami memperkenalkan kawalan kemas kini di mana purata tenaga puncak-ke-korelasi (APCE) dan jarak antara lokasi objek yang diperolehi oleh ciri HOG dan ciri buatan tangan digunakan untuk mengesan keabnormalan keadaan di sekeliling objek. APCE dan jarak menunjukkan kebolehpercayaan tindak balas penapis, oleh itu jika keabnormalan dikesan, kaedah yang dicadangkan tidak mengemas kini skala dan lokasi objek yang dianggarkan oleh tindak balas penapis. Dalam percubaan, penjejak kami (STPSE) mencapai prestasi ketara dan masa nyata dengan hanya CPU untuk jujukan penanda aras yang mencabar (OTB2013, OTB2015 dan TC128).
Zhaoqian TANG
Meiji University
Kaoru ARAKAWA
Meiji University
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.
Salinan
Zhaoqian TANG, Kaoru ARAKAWA, "Spatial-Temporal Regularized Correlation Filter with Precise State Estimation for Visual Tracking" in IEICE TRANSACTIONS on Fundamentals,
vol. E105-A, no. 6, pp. 914-922, June 2022, doi: 10.1587/transfun.2021EAP1087.
Abstract: Recently, the performances of discriminative correlation filter (CF) trackers are getting better and better in visual tracking. In this paper, we propose spatial-temporal regularization with precise state estimation based on discriminative correlation filter (STPSE) in order to achieve more significant tracking performance. First, we consider the continuous change of the object state, using the information from the previous two filters for training the correlation filter model. Here, we train the correlation filter model with the hand-crafted features. Second, we introduce update control in which average peak-to-correlation energy (APCE) and the distance between the object locations obtained by HOG features and hand-crafted features are utilized to detect abnormality of the state around the object. APCE and the distance indicate the reliability of the filter response, thus if abnormality is detected, the proposed method does not update the scale and the object location estimated by the filter response. In the experiment, our tracker (STPSE) achieves significant and real-time performance with only CPU for the challenging benchmark sequence (OTB2013, OTB2015, and TC128).
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.2021EAP1087/_p
Salinan
@ARTICLE{e105-a_6_914,
author={Zhaoqian TANG, Kaoru ARAKAWA, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Spatial-Temporal Regularized Correlation Filter with Precise State Estimation for Visual Tracking},
year={2022},
volume={E105-A},
number={6},
pages={914-922},
abstract={Recently, the performances of discriminative correlation filter (CF) trackers are getting better and better in visual tracking. In this paper, we propose spatial-temporal regularization with precise state estimation based on discriminative correlation filter (STPSE) in order to achieve more significant tracking performance. First, we consider the continuous change of the object state, using the information from the previous two filters for training the correlation filter model. Here, we train the correlation filter model with the hand-crafted features. Second, we introduce update control in which average peak-to-correlation energy (APCE) and the distance between the object locations obtained by HOG features and hand-crafted features are utilized to detect abnormality of the state around the object. APCE and the distance indicate the reliability of the filter response, thus if abnormality is detected, the proposed method does not update the scale and the object location estimated by the filter response. In the experiment, our tracker (STPSE) achieves significant and real-time performance with only CPU for the challenging benchmark sequence (OTB2013, OTB2015, and TC128).},
keywords={},
doi={10.1587/transfun.2021EAP1087},
ISSN={1745-1337},
month={June},}
Salinan
TY - JOUR
TI - Spatial-Temporal Regularized Correlation Filter with Precise State Estimation for Visual Tracking
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 914
EP - 922
AU - Zhaoqian TANG
AU - Kaoru ARAKAWA
PY - 2022
DO - 10.1587/transfun.2021EAP1087
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E105-A
IS - 6
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - June 2022
AB - Recently, the performances of discriminative correlation filter (CF) trackers are getting better and better in visual tracking. In this paper, we propose spatial-temporal regularization with precise state estimation based on discriminative correlation filter (STPSE) in order to achieve more significant tracking performance. First, we consider the continuous change of the object state, using the information from the previous two filters for training the correlation filter model. Here, we train the correlation filter model with the hand-crafted features. Second, we introduce update control in which average peak-to-correlation energy (APCE) and the distance between the object locations obtained by HOG features and hand-crafted features are utilized to detect abnormality of the state around the object. APCE and the distance indicate the reliability of the filter response, thus if abnormality is detected, the proposed method does not update the scale and the object location estimated by the filter response. In the experiment, our tracker (STPSE) achieves significant and real-time performance with only CPU for the challenging benchmark sequence (OTB2013, OTB2015, and TC128).
ER -