Real-Time ‘Actor-Critic’ Tracking

Chen, Bo‐Yu; Wang, Dong; Li, Peixia; Wang, Shuang; Lu, Huchuan

doi:10.1007/978-3-030-01234-2_20

Cited by 139 publications

(86 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…EAST [22] 0.638 -0.629 159 PTAV [12] 0.663 0.581 0.635 25 ACT [5] 0.657 -0.625 30 RT-MDNet [24] --0.650 46 Ours 0.693 0.653 0.687 120 Table 3. Comparison with state-of-the-art real-time trackers on OTB dataset.…”

Section: Siamfc-based Trackersmentioning

confidence: 99%

SPM-Tracker: Series-Parallel Matching for Real-Time Visual Object Tracking

Wang

Luo

Xiong

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

231

135

View full text Add to dashboard Cite

The greatest challenge facing visual object tracking is the simultaneous requirements on robustness and discrimination power. In this paper, we propose a SiamFC-based tracker, named SPM-Tracker, to tackle this challenge. The basic idea is to address the two requirements in two separate matching stages. Robustness is strengthened in the coarse matching (CM) stage through generalized training while discrimination power is enhanced in the fine matching (FM) stage through a distance learning network. The two stages are connected in series as the input proposals of the FM stage are generated by the CM stage. They are also connected in parallel as the matching scores and box location refinements are fused to generate the final results. This innovative series-parallel structure takes advantage of both stages and results in superior performance. The proposed SPM-Tracker, running at 120fps on GPU, achieves an AUC of 0.687 on OTB-100 and an EAO of 0.434 on VOT-16, exceeding other real-time trackers by a notable margin.

show abstract

Section: Siamfc-based Trackersmentioning

confidence: 99%

SPM-Tracker: Series-Parallel Matching for Real-Time Visual Object Tracking

Wang

Luo

Xiong

et al. 2019

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

231

135

View full text Add to dashboard Cite

show abstract

“…Whereas the success plot reports the percentages of frames where the overlap between the predicted and the ground truth bounding boxes is higher than a series of given ratios. We compare our algorithm with twelve state-of-theart trackers including nine real-time deep trackers (ACT [5], StructSiam [44], SiamRPN [22], ECO-HC [7], PTAV [13], CFNet [35], Dsiam [16], LCT [27], SiameFC [3]) and three traditional trackers (Staple [2], DSST [8], KCF [18]). Figure 6 illustrates the precision and success plots of all compared trackers over OTB-2015, which shows the proposed tracker achieves very good performance (merely a slightly lower than ECO-HC in success).…”

Section: Evaluation On the Otb-2015 Datasetmentioning

confidence: 99%

“…We also adopt both success and precision plots to evaluate different trackers (the same evaluation protocol as OTB-2015). We compare our algorithm with eleven trackers, including ACT [5], PTAV [13], Dsiam [16], SiameFC [3], HCFT [26], FCNT [36], STCT [37], BACF [15], SRDCF [9], KCF [18] and MEEM [41]. Figure 7 shows that our tracker achieves the best results in terms of both precision and success criterion.…”

Section: Evaluation On the Tc-128 Datasetmentioning

confidence: 99%

GradNet: Gradient-Guided Network for Visual Object Tracking

Chen

Ouyang

et al. 2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

Self Cite

281

122

View full text Add to dashboard Cite

The fully-convolutional siamese network based on template matching has shown great potentials in visual tracking. During testing, the template is fixed with the initial target feature and the performance totally relies on the general matching ability of the siamese network. However, this manner cannot capture the temporal variations of targets or background clutter. In this work, we propose a novel gradient-guided network to exploit the discriminative information in gradients and update the template in the siamese network through feed-forward and backward operations. To be specific, the algorithm can utilize the information from the gradient to update the template in the current frame. In addition, a template generalization training method is proposed to better use gradient information and avoid overfitting. To our knowledge, this work is the first attempt to exploit the information in the gradient for template update in siamese-based trackers. Extensive experiments on recent benchmarks demonstrate that our method achieves better performance than other state-of-the-art trackers. The source codes are available at

show abstract

“…In [39], authors presented a tracker which, at every time step, decides to shift the current bounding box while remaining on the same frame, to stop the shift process and move to the next frame, to update on-line the weights of the model or to re-initialize the tracker if the target is considered lost. Finally, [3] proposed to substitute the discrete action framework of [52] with continuous actions, thus performing just a single action at every frame.…”

Section: Deep Rl For Visual Trackingmentioning

confidence: 99%

“…In the t-th frame, the agent is provided with the state s t and outputs the continuous action a t which consists in the relative motion of the target object, i.e. it indicates how its bounding box, which is known in frame t − 1, should move to enclose the target in the frame t. This approach is similar to the MDP formulation given by Chen et al [3], however we propose different definitions for the states, actions and rewards.…”

Section: Problem Settingmentioning

confidence: 99%

Visual Tracking by Means of Deep Reinforcement Learning and an Expert Demonstrator

Dunnhofer

Martinel

Foresti

et al. 2019

2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)

View full text Add to dashboard Cite

In the last decade many different algorithms have been proposed to track a generic object in videos. Their execution on recent large-scale video datasets can produce a great amount of various tracking behaviours. New trends in Reinforcement Learning showed that demonstrations of an expert agent can be efficiently used to speed-up the process of policy learning. Taking inspiration from such works and from the recent applications of Reinforcement Learning to visual tracking, we propose two novel trackers, A3CT, which exploits demonstrations of a state-of-the-art tracker to learn an effective tracking policy, and A3CTD, that takes advantage of the same expert tracker to correct its behaviour during tracking. Through an extensive experimental validation on the GOT-10k, OTB-100, LaSOT, UAV123 and VOT benchmarks, we show that the proposed trackers achieve state-of-the-art performance while running in real-time. arXiv:1909.08487v1 [cs.CV] 18 Sep 2019 =

show abstract

Real-Time ‘Actor-Critic’ Tracking

Cited by 139 publications

References 48 publications

SPM-Tracker: Series-Parallel Matching for Real-Time Visual Object Tracking

SPM-Tracker: Series-Parallel Matching for Real-Time Visual Object Tracking

GradNet: Gradient-Guided Network for Visual Object Tracking

Visual Tracking by Means of Deep Reinforcement Learning and an Expert Demonstrator

Contact Info

Product

Resources

About