Deep neural networks have achieved great success in visual tracking by learning a generic representation and leveraging large amounts of training data to improve performance. Most generic object trackers are trained from scratch online and do not benefit from a large number of videos available for offline training. We present a real-time generic object tracker capable of incorporating temporal information into its model, learning from many examples offline and quickly updating online. During the training process, the pre-trained weight of convolution layer is updated lagging behind, and the input video sequence length is gradually increased for fast convergence. Furthermore, only the hidden states in recurrent network are updated to guarantee the real-time tracking speed. The experimental results show that the proposed tracking method is capable of tracking objects at 150 fps with higher predicting overlap rate, and achieves more robustness in multiple benchmarks than state-of-the-art performance.
Rui CHEN
Nanjing Institute of Technology
Ying TONG
Nanjing Institute of Technology
Ruiyu LIANG
Nanjing Institute of Technology
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.
Copy
Rui CHEN, Ying TONG, Ruiyu LIANG, "Real-Time Generic Object Tracking via Recurrent Regression Network" in IEICE TRANSACTIONS on Information,
vol. E103-D, no. 3, pp. 602-611, March 2020, doi: 10.1587/transinf.2019EDP7240.
Abstract: Deep neural networks have achieved great success in visual tracking by learning a generic representation and leveraging large amounts of training data to improve performance. Most generic object trackers are trained from scratch online and do not benefit from a large number of videos available for offline training. We present a real-time generic object tracker capable of incorporating temporal information into its model, learning from many examples offline and quickly updating online. During the training process, the pre-trained weight of convolution layer is updated lagging behind, and the input video sequence length is gradually increased for fast convergence. Furthermore, only the hidden states in recurrent network are updated to guarantee the real-time tracking speed. The experimental results show that the proposed tracking method is capable of tracking objects at 150 fps with higher predicting overlap rate, and achieves more robustness in multiple benchmarks than state-of-the-art performance.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2019EDP7240/_p
Copy
@ARTICLE{e103-d_3_602,
author={Rui CHEN, Ying TONG, Ruiyu LIANG, },
journal={IEICE TRANSACTIONS on Information},
title={Real-Time Generic Object Tracking via Recurrent Regression Network},
year={2020},
volume={E103-D},
number={3},
pages={602-611},
abstract={Deep neural networks have achieved great success in visual tracking by learning a generic representation and leveraging large amounts of training data to improve performance. Most generic object trackers are trained from scratch online and do not benefit from a large number of videos available for offline training. We present a real-time generic object tracker capable of incorporating temporal information into its model, learning from many examples offline and quickly updating online. During the training process, the pre-trained weight of convolution layer is updated lagging behind, and the input video sequence length is gradually increased for fast convergence. Furthermore, only the hidden states in recurrent network are updated to guarantee the real-time tracking speed. The experimental results show that the proposed tracking method is capable of tracking objects at 150 fps with higher predicting overlap rate, and achieves more robustness in multiple benchmarks than state-of-the-art performance.},
keywords={},
doi={10.1587/transinf.2019EDP7240},
ISSN={1745-1361},
month={March},}
Copy
TY - JOUR
TI - Real-Time Generic Object Tracking via Recurrent Regression Network
T2 - IEICE TRANSACTIONS on Information
SP - 602
EP - 611
AU - Rui CHEN
AU - Ying TONG
AU - Ruiyu LIANG
PY - 2020
DO - 10.1587/transinf.2019EDP7240
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E103-D
IS - 3
JA - IEICE TRANSACTIONS on Information
Y1 - March 2020
AB - Deep neural networks have achieved great success in visual tracking by learning a generic representation and leveraging large amounts of training data to improve performance. Most generic object trackers are trained from scratch online and do not benefit from a large number of videos available for offline training. We present a real-time generic object tracker capable of incorporating temporal information into its model, learning from many examples offline and quickly updating online. During the training process, the pre-trained weight of convolution layer is updated lagging behind, and the input video sequence length is gradually increased for fast convergence. Furthermore, only the hidden states in recurrent network are updated to guarantee the real-time tracking speed. The experimental results show that the proposed tracking method is capable of tracking objects at 150 fps with higher predicting overlap rate, and achieves more robustness in multiple benchmarks than state-of-the-art performance.
ER -