Distractor-Aware Siamese Networks for Visual Object Tracking

Zhu, Zheng; Wang, Qiang; Li, Bo; Wu, Wei; Yan, Junjie; Hu, Weiming

doi:10.1007/978-3-030-01240-3_7

Cited by 1,259 publications

(941 citation statements)

References 41 publications

Supporting

Mentioning

849

Contrasting

Unclassified

Order By: Relevance

“…SiameseRPN [39] proposes an offline trained Siamese Region Proposal Network (RPN). DaSiameseRPN [42] improves SiameseRPN by introducing a distractor-aware module. C-RPN [43] proposes Siamese cascaded RPNs to solve the problem of class imbalance by performing hard negative sampling.…”

Section: Related Workmentioning

confidence: 99%

Object-adaptive LSTM network for real-time visual tracking with adversarial data augmentation

Yan

Chen

et al. 2020

Neurocomputing

View full text Add to dashboard Cite

In recent years, deep learning based visual tracking methods have obtained great success owing to the powerful feature representation ability of Convolutional Neural Networks (CNNs). Among these methods, classification-based tracking methods exhibit excellent performance while their speeds are heavily limited by the expensive computation for massive proposal feature extraction. In contrast, matching-based tracking methods (such as Siamese networks) possess remarkable speed superiority.However, the absence of online updating renders these methods unadaptable to significant object appearance variations. In this paper, we propose a novel real-time visual tracking method, which adopts an object-adaptive LSTM network to effectively capture the video sequential dependencies and adaptively learn the object appearance variations. For high computational efficiency, we also present a fast proposal selection strategy, which utilizes the matching-based tracking method to pre-estimate dense proposals and selects high-quality ones to feed to the LSTM network for classification. This strategy efficiently filters out some irrelevant proposals and avoids the redundant computation for feature extraction, which enables our method to operate faster than conventional classification-based tracking methods. In addition, to handle the problems of sample inadequacy and class imbalance during online tracking, we adopt a data augmentation technique based on the Generative Adversarial Network (GAN) to facilitate the training of the LSTM network. Extensive experiments on four visual tracking benchmarks demonstrate the state-of-the-art performance of our method in terms of both tracking accuracy and speed, which exhibits great potentials of recurrent structures for visual tracking.

show abstract

Section: Related Workmentioning

confidence: 99%

Object-adaptive LSTM network for real-time visual tracking with adversarial data augmentation

Yan

Chen

et al. 2020

Neurocomputing

View full text Add to dashboard Cite

show abstract

“…If an IoU between an estimated bounding box and its previous bounding box, is under a threshold µ, we treat this situation as a tracking failure. And we use the DaSiamRPN tracker [41] to track the target object again on the reconstructed intensity event frames for recovering the tracking status from the tracking failure.…”

Section: Event-based Tracking-by-detectionmentioning

confidence: 99%

“…ECO [10] is a state-of-the-art object tracking method that employs compact samples to train continuous convolutional operators for visual tracking. DaSiamRPN [41] is a stateof-the-art tracking method that improves the fully-convolutional siamese network by using distractor-aware training. DaSiamRPN-E, which uses ATSLTD frames as its input, is an event-based variant of DaSiamRPN.…”

Section: Experiments 41 Experimental Settingsmentioning

confidence: 99%

Asynchronous Tracking-by-Detection on Adaptive Time Surfaces for Event-based Object Tracking

Chen

Liang

et al. 2019

Proceedings of the 27th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Event cameras, which are asynchronous bio-inspired vision sensors, have shown great potential in a variety of situations, such as fast motion and low illumination scenes. However, most of the event-based object tracking methods are designed for scenarios with untextured objects and uncluttered backgrounds. There are few event-based object tracking methods that support bounding box-based object tracking. The main idea behind this work is to propose an asynchronous Event-based Tracking-by-Detection (ETD) method for generic bounding box-based object tracking. To achieve this goal, we present an Adaptive Time-Surface with Linear Time Decay (AT-SLTD) event-to-frame conversion algorithm, which asynchronously and effectively warps the spatio-temporal information of asynchronous retinal events to a sequence of ATSLTD frames with clear object contours. We feed the sequence of ATSLTD frames to the proposed ETD method to perform accurate and efficient object tracking, which leverages the high temporal resolution property of event cameras. We compare the proposed ETD method with seven popular object tracking methods, that are based on conventional cameras or event cameras, and two variants of ETD. The experimental results show the superiority of the proposed ETD method in handling various challenging environments.

show abstract

“…Other variants of the Siamese FC tracker have outperformed the baseline SiamFC [1] tracker like SiamRPN [8], SA-Siam [6], DaSiamRPN [18] and MEMTrack [16]. They have proposed various improvements to the original SiamFC tracker such as attention mechanims, region proposal, etc.…”

Section: Tracking Objects In Video Surveillancementioning

confidence: 99%

“…Hence, when appearance changes abruptly or the object is occluded or partially leaves the search region, the SiamFC tracker temporarily drifts to a location that has a high response map score. Recently, DaSi-amRPN [18] tracker based on the SiamFC tracking technique further improved by incorporating distracter awareness has produced state of the art results on various tracking benchmarks.…”

Section: Introductionmentioning

confidence: 99%

On the Interaction Between Deep Detectors and Siamese Trackers in Video Surveillance

Kiran

Tiwari

Nguyen-Meidine

et al. 2019

2019 16th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)

View full text Add to dashboard Cite

Visual object tracking is an important function in many real-time video surveillance applications, such as localization and spatio-temporal recognition of persons. In realworld applications, an object detector and tracker must interact on a periodic basis to discover new objects, and thereby to initiate tracks. Periodic interactions with the detector can also allow the tracker to validate and/or update its object template with new bounding boxes. However, bounding boxes provided by a state-of-the-art detector are noisy, due to changes in appearance, background and occlusion, which can cause the tracker to drift. Moreover, CNN-based detectors can provide a high level of accuracy at the expense of computational complexity, so interactions should be minimized for real-time applications. In this paper, a new approach is proposed to manage detectortracker interactions for trackers from the Siamese-FC family. By integrating a change detection mechanism into a deep Siamese-FC tracker, its template can be adapted in response to changes in a target's appearance that lead to drifts during tracking. An abrupt change detection triggers an update of tracker template using the bounding box produced by the detector, while in the case of a gradual change, the detector is used to update an evolving set of templates for robust matching. Experiments were performed using state-of-the-art Siamese-FC trackers and the YOLOv3 detector on a subset of videos from the OTB-100 dataset that mimic video surveillance scenarios. Results highlight the importance for reliable VOT of using accurate detectors. They also indicate that our adaptive Siamese trackers are robust to noisy object detections, and can significantly improve the performance of Siamese-FC tracking.

show abstract

Distractor-Aware Siamese Networks for Visual Object Tracking

Cited by 1,259 publications

References 41 publications

Object-adaptive LSTM network for real-time visual tracking with adversarial data augmentation

Object-adaptive LSTM network for real-time visual tracking with adversarial data augmentation

Asynchronous Tracking-by-Detection on Adaptive Time Surfaces for Event-based Object Tracking

On the Interaction Between Deep Detectors and Siamese Trackers in Video Surveillance

Contact Info

Product

Resources

About