Tracking by Instance Detection: A Meta-Learning Approach

Wang, Guangting; Luo, Chong; Sun, Xiaoyan; Xiong, Zhiwei; Zeng, Wenjun

doi:10.1109/cvpr42600.2020.00632

Cited by 191 publications

(92 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In terms of tracking speed, we compared it with night state-of-the-art trackers including SiamBAN [ 6 ], PrDiMP [ 18 ], Retina-MAML [ 39 ], FCOS-MAML [ 39 ], SiamRPN++ [ 7 ], SiamFC++ [ 27 ], ATOM [ 16 ], and SPS [ 36 ]. Our TA-Siam achieves the advanced EAO score (0.469) while running at 45 FPS.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Robust Template Adjustment Siamese Network for Object Visual Tracking

Tang

Qin

Zhang

2021

Sensors

View full text Add to dashboard Cite

Most of the existing trackers address the visual tracking problem by extracting an appearance template from the first frame, which is used to localize the target in the current frame. Unfortunately, they typically face the model degeneration challenge, which easily results in model drift and target loss. To address this issue, a novel Template Adjustment Siamese Network (TA-Siam) is proposed in this paper. The proposed framework TA-Siam consists of two simple subnetworks: The template adjustment subnetwork for feature extraction and the classification-regression subnetwork for bounding box prediction. The template adjustment module adaptively uses the feature of subsequent frames to adjust the current template. It makes the template adapt to the target appearance variation of long-term sequence and effectively overcomes model drift problem of Siamese networks. In order to reduce classification errors, the rhombus labels are proposed in our TA-Siam. For more efficient learning and faster convergence, our proposed tracker uses a more effective regression loss in the training process. Extensive experiments and comparisons with trackers are conducted on the challenging benchmarks including VOT2016, VOT2018, OTB50, OTB100, GOT-10K, and LaSOT. Our TA-Siam achieves state-of-the-art performance at the speed of 45 FPS.

show abstract

Section: Methodsmentioning

confidence: 99%

“… The “FCOS-MAML” and “Retina-MAML” indicate MAML [ 39 ] trackers based on detectors RetinaNet [ 29 ] and FCOS [ 26 ] respectively. ↑ indicates that the larger the value, the better the performance.…”

Section: Figurementioning

confidence: 99%

Robust Template Adjustment Siamese Network for Object Visual Tracking

Tang

Qin

Zhang

2021

Sensors

View full text Add to dashboard Cite

show abstract

“…MemTrack [197] 0.849 0.642 DSLT [169] 0.934 0.683 SiamFC [198] 0.809 0.607 StructSiam [165] 0.880 0.638 SSD [160] 0.813 0.637 CR-RE [199] 0.677 0.538 HCFT [200] 0.923 0.638 EMDSLT [166] 0.853 0.626 MDSLT [166] 0.815 0.600…”

Section: Trackers Precision Successmentioning

confidence: 99%

Advances in Deep Learning Methods for Visual Tracking: Literature Review and Fundamentals

Zhang

Jiang

Fan

et al. 2021

Int. J. Autom. Comput.

View full text Add to dashboard Cite

Recently, deep learning has achieved great success in visual tracking tasks, particularly in single-object tracking. This paper provides a comprehensive review of state-of-the-art single-object tracking algorithms based on deep learning. First, we introduce basic knowledge of deep visual tracking, including fundamental concepts, existing algorithms, and previous reviews. Second, we briefly review existing deep learning methods by categorizing them into data-invariant and data-adaptive methods based on whether they can dynamically change their model parameters or architectures. Then, we conclude with the general components of deep trackers. In this way, we systematically analyze the novelties of several recently proposed deep trackers. Thereafter, popular datasets such as Object Tracking Benchmark (OTB) and Visual Object Tracking (VOT) are discussed, along with the performances of several deep trackers. Finally, based on observations and experimental results, we discuss three different characteristics of deep trackers, i.e., the relationships between their general components, exploration of more effective tracking frameworks, and interpretability of their motion estimation components.

show abstract

“…The major benefit of meta-learning over conventional transfer learning (or fine-tuning) approaches is that it allows the network to utilize its pretrained weights to effectively predict the unseen examples of the new underlying task without having to retrain on the large (and diverse) set of training examples for this current task to avoid overfitting [ 17 ]. Meta-learning has not only been employed for the supervised classification [ 16 ] and detection [ 21 ] tasks. It has also been used to acquire unlabeled data representation in an unsupervised manner [ 22 ].…”

Section: Related Workmentioning

confidence: 99%

Meta-Transfer Learning Driven Tensor-Shot Detector for the Autonomous Localization and Recognition of Concealed Baggage Threats

Hassan

Shafay

Akçay

et al. 2020

Sensors

View full text Add to dashboard Cite

Screening baggage against potential threats has become one of the prime aviation security concerns all over the world, where manual detection of prohibited items is a time-consuming and hectic process. Many researchers have developed autonomous systems to recognize baggage threats using security X-ray scans. However, all of these frameworks are vulnerable against screening cluttered and concealed contraband items. Furthermore, to the best of our knowledge, no framework possesses the capacity to recognize baggage threats across multiple scanner specifications without an explicit retraining process. To overcome this, we present a novel meta-transfer learning-driven tensor-shot detector that decomposes the candidate scan into dual-energy tensors and employs a meta-one-shot classification backbone to recognize and localize the cluttered baggage threats. In addition, the proposed detection framework can be well-generalized to multiple scanner specifications due to its capacity to generate object proposals from the unified tensor maps rather than diversified raw scans. We have rigorously evaluated the proposed tensor-shot detector on the publicly available SIXray and GDXray datasets (containing a cumulative of 1,067,381 grayscale and colored baggage X-ray scans). On the SIXray dataset, the proposed framework achieved a mean average precision (mAP) of 0.6457, and on the GDXray dataset, it achieved the precision and F1 score of 0.9441 and 0.9598, respectively. Furthermore, it outperforms state-of-the-art frameworks by 8.03% in terms of mAP, 1.49% in terms of precision, and 0.573% in terms of F1 on the SIXray and GDXray dataset, respectively.

show abstract

Tracking by Instance Detection: A Meta-Learning Approach

Cited by 191 publications

References 39 publications

Robust Template Adjustment Siamese Network for Object Visual Tracking

Robust Template Adjustment Siamese Network for Object Visual Tracking

Advances in Deep Learning Methods for Visual Tracking: Literature Review and Fundamentals

Meta-Transfer Learning Driven Tensor-Shot Detector for the Autonomous Localization and Recognition of Concealed Baggage Threats

Contact Info

Product

Resources

About