TRBACF: Learning temporal regularized correlation filters for high performance online visual object tracking

Yuan, Di; Shu, Xiu; He, Zhenyu

doi:10.1016/j.jvcir.2020.102882

Cited by 37 publications

(18 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the future work, more strict temporal constraint can be adopted to overcome this limitation. Another future work is to construct a low-complexity framework to solve the optimization problem [42][43][44][45][46].…”

Section: Discussionmentioning

confidence: 99%

Depth Trajectory-Aware Stereoscopic Video Retargeting

Wang

Shao

2021

IEEE Access

View full text Add to dashboard Cite

The image/video retargeting (resolution adaptation) problem-adaption of source data (image/video) to different screen resolution-has garnered wide attention due to compelling applications in intelligent adaptive display. Most of the current image/video displaying application achieves the adaption through scaling source uniformly. Compared with traditional image/video retargeting, stereoscopic video retargeting poses new challenges that should preserve shape and depth of salient objects with temporal coherence. In this paper, we propose a depth trajectory-aware stereoscopic video retargeting method by imposing constraints to optimize the coordinates and depths simultaneously. To reduce visual discomfort induced by fast depth motion, our method uses a temporal depth distortion energy to optimize the depth trajectory in temporal direction. As a result, our solution can preserve shape, depth and temporal fidelity of salient objects simultaneously with comfortable visual experience.

show abstract

Section: Discussionmentioning

confidence: 99%

Depth Trajectory-Aware Stereoscopic Video Retargeting

Wang

Shao

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…This tracker is equipped with deep neural network features and the ability of joint feature selection and filter learning. The TRBACF [47] tracker designs a temporal regularization strategy which can efficiently adjust the model to adapt to the change of the tracking scenes, and this makes it more robust to complex environments. The ARCF [48] filter focuses on addressing the boundary effect problem in a correlation filter and adds restrictions to the alteration rate in response maps, so aberrances in detection can be largely suppressed, which makes the tracker more robust and accurate.…”

Section: Related Workmentioning

confidence: 99%

MFCFSiam: A Correlation-Filter-Guided Siamese Network with Multifeature for Visual Tracking

Xing

et al. 2020

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

With the development of deep learning, trackers based on convolutional neural networks (CNNs) have made significant achievements in visual tracking over the years. The fully connected Siamese network (SiamFC) is a typical representation of those trackers. SiamFC designs a two-branch architecture of a CNN and models’ visual tracking as a general similarity-learning problem. However, the feature maps it uses for visual tracking are only from the last layer of the CNN. Those features contain high-level semantic information but lack sufficiently detailed texture information. This means that the SiamFC tracker tends to drift when there are other same-category objects or when the contrast between the target and the background is very low. Focusing on addressing this problem, we design a novel tracking algorithm that combines a correlation filter tracker and the SiamFC tracker into one framework. In this framework, the correlation filter tracker can use the Histograms of Oriented Gradients (HOG) and color name (CN) features to guide the SiamFC tracker. This framework also contains an evaluation criterion which we design to evaluate the tracking result of the two trackers. If this criterion finds the SiamFC tracker fails in some cases, our framework will use the tracking result from the correlation filter tracker to correct the SiamFC. In this way, the defects of SiamFC’s high-level semantic features are remedied by the HOG and CN features. So, our algorithm provides a framework which combines two trackers together and makes them complement each other in visual tracking. And to the best of our knowledge, our algorithm is also the first one which designs an evaluation criterion using correlation filter and zero padding to evaluate the tracking result. Comprehensive experiments are conducted on the Online Tracking Benchmark (OTB), Temple Color (TC128), Benchmark for UAV Tracking (UAV-123), and Visual Object Tracking (VOT) Benchmark. The results show that our algorithm achieves quite a competitive performance when compared with the baseline tracker and several other state-of-the-art trackers.

show abstract

“…Correlation operation is computed in the Fast Fourier Transform (FFT) of Fourier domain to accelerate the speed of calculation [8], but the periodic assumption of Fourier Transform leads to boundary effect. Many researchers [9,10,11,12,13] focus on introducing the regularization item to compensate filter coefficient in the learning process through the regularization weight, so as to enhance the performance of tracking filters. Regardless of boundary effect, several trackers [9, 10, 14] exploit some powerful and complicated deep features (features extracted from trained deep network) and merge them with handcraft features.…”

Section: Introductionmentioning

confidence: 99%

Multi-Model and Multi-Expert Correlation Filter for High-Speed Tracking

Liang

Wang

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Tracking algorithm based on correlation filter have been extensively investigated due to their powerful performance in benchmark datasets and competitions. However, the periodic assumption has contributed boundary effects and the complex scenarios will give rise to model drift, which have an extremely negative effect on both tracking precision and success rate. To mitigate these challenges, a novel multi-model and multi-expert correlation filter (MMCF) approach is proposed in this paper. The key innovation of the proposed method is to employ multiple models and experts for tracking. Multiple models can excavate diverse a large quantity of feature information, and the target information between different models can complement each other. Multiple experts provide several possible predictions, while an evaluation mechanism selects the most reliable prediction utilizing their assessment scores. To further improve the performance, an adaption strategy is utilized to update the multiple models, which can reduce the weight of the bad samples to prevent model drift. Experiments performed on three recent benchmark datasets OTB50, OTB2013, OTB100, TC-128 and UAV123@10fps, have demonstrated the superiority of our approach in comparison to the state-of-the-art trackers. Our MMCF tracker operates a speed of about 58 frames per second (FPS) running on a single central processing unit(CPU). INDEX TERMS Correlation filter; boundary effect, model drift, multiple models, multiple experts.

show abstract

TRBACF: Learning temporal regularized correlation filters for high performance online visual object tracking

Cited by 37 publications

References 25 publications

Depth Trajectory-Aware Stereoscopic Video Retargeting

Depth Trajectory-Aware Stereoscopic Video Retargeting

MFCFSiam: A Correlation-Filter-Guided Siamese Network with Multifeature for Visual Tracking

Multi-Model and Multi-Expert Correlation Filter for High-Speed Tracking

Contact Info

Product

Resources

About