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
Embora os rastreadores baseados em filtros de correlação tenham demonstrado excelente desempenho para rastreamento visual de objetos, ainda existem vários desafios a serem enfrentados. Neste trabalho, propomos um novo rastreador baseado na estrutura do filtro de correlação. Os rastreadores tradicionais enfrentam dificuldade em se adaptar com precisão às mudanças na escala do alvo quando o alvo se move rapidamente. Para resolver isso, sugerimos um esquema adaptativo de escala baseado em escalas de predição. Também incorporamos um método de atualização de modelo adaptativo baseado em velocidade para melhorar ainda mais o desempenho geral de rastreamento. Experimentos com amostras dos conjuntos de dados OTB100 e KITTI demonstram que nosso método supera os algoritmos de rastreamento de última geração existentes em cenas de movimento rápido.
Zuopeng ZHAO
CUMT,Mine Digitization Engineering Research Center of Minstry of Education of the People's Republic of China
Hongda ZHANG
CUMT,Mine Digitization Engineering Research Center of Minstry of Education of the People's Republic of China
Yi LIU
CUMT,Mine Digitization Engineering Research Center of Minstry of Education of the People's Republic of China
Nana ZHOU
CUMT,Mine Digitization Engineering Research Center of Minstry of Education of the People's Republic of China
Han ZHENG
CUMT,Mine Digitization Engineering Research Center of Minstry of Education of the People's Republic of China
Shanyi SUN
CUMT,Mine Digitization Engineering Research Center of Minstry of Education of the People's Republic of China
Xiaoman LI
CUMT,Mine Digitization Engineering Research Center of Minstry of Education of the People's Republic of China
Sili XIA
CUMT,Mine Digitization Engineering Research Center of Minstry of Education of the People's Republic of China
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.
Copiar
Zuopeng ZHAO, Hongda ZHANG, Yi LIU, Nana ZHOU, Han ZHENG, Shanyi SUN, Xiaoman LI, Sili XIA, "Prediction-Based Scale Adaptive Correlation Filter Tracker" in IEICE TRANSACTIONS on Information,
vol. E102-D, no. 11, pp. 2267-2271, November 2019, doi: 10.1587/transinf.2019EDL8101.
Abstract: Although correlation filter-based trackers have demonstrated excellent performance for visual object tracking, there remain several challenges to be addressed. In this work, we propose a novel tracker based on the correlation filter framework. Traditional trackers face difficulty in accurately adapting to changes in the scale of the target when the target moves quickly. To address this, we suggest a scale adaptive scheme based on prediction scales. We also incorporate a speed-based adaptive model update method to further improve overall tracking performance. Experiments with samples from the OTB100 and KITTI datasets demonstrate that our method outperforms existing state-of-the-art tracking algorithms in fast motion scenes.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2019EDL8101/_p
Copiar
@ARTICLE{e102-d_11_2267,
author={Zuopeng ZHAO, Hongda ZHANG, Yi LIU, Nana ZHOU, Han ZHENG, Shanyi SUN, Xiaoman LI, Sili XIA, },
journal={IEICE TRANSACTIONS on Information},
title={Prediction-Based Scale Adaptive Correlation Filter Tracker},
year={2019},
volume={E102-D},
number={11},
pages={2267-2271},
abstract={Although correlation filter-based trackers have demonstrated excellent performance for visual object tracking, there remain several challenges to be addressed. In this work, we propose a novel tracker based on the correlation filter framework. Traditional trackers face difficulty in accurately adapting to changes in the scale of the target when the target moves quickly. To address this, we suggest a scale adaptive scheme based on prediction scales. We also incorporate a speed-based adaptive model update method to further improve overall tracking performance. Experiments with samples from the OTB100 and KITTI datasets demonstrate that our method outperforms existing state-of-the-art tracking algorithms in fast motion scenes.},
keywords={},
doi={10.1587/transinf.2019EDL8101},
ISSN={1745-1361},
month={November},}
Copiar
TY - JOUR
TI - Prediction-Based Scale Adaptive Correlation Filter Tracker
T2 - IEICE TRANSACTIONS on Information
SP - 2267
EP - 2271
AU - Zuopeng ZHAO
AU - Hongda ZHANG
AU - Yi LIU
AU - Nana ZHOU
AU - Han ZHENG
AU - Shanyi SUN
AU - Xiaoman LI
AU - Sili XIA
PY - 2019
DO - 10.1587/transinf.2019EDL8101
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E102-D
IS - 11
JA - IEICE TRANSACTIONS on Information
Y1 - November 2019
AB - Although correlation filter-based trackers have demonstrated excellent performance for visual object tracking, there remain several challenges to be addressed. In this work, we propose a novel tracker based on the correlation filter framework. Traditional trackers face difficulty in accurately adapting to changes in the scale of the target when the target moves quickly. To address this, we suggest a scale adaptive scheme based on prediction scales. We also incorporate a speed-based adaptive model update method to further improve overall tracking performance. Experiments with samples from the OTB100 and KITTI datasets demonstrate that our method outperforms existing state-of-the-art tracking algorithms in fast motion scenes.
ER -