Learning to Filter: Siamese Relation Network for Robust Tracking

Cheng, Siyuan; Zhong, Bineng; Li, Guorong; Liu, Xin; Tang, Zhenjun; Li, Xianxian; Wang, Jing

doi:10.1109/cvpr46437.2021.00440

Cited by 114 publications

(40 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As shown in Table 1, our proposed SiamRPN++-RBO, SiamBAN-RBO and SiamPW-RBO achieve the AUC scores of 69.9%, 70.1% and 69.8%, respectively. Compared with the recent proposed Siamese trackers such as SiamRN [9] and SiamGAT [18], our three trackers achieve better or competitive performance against them.…”

Section: Comparison With State-of-the-art Trackersmentioning

confidence: 87%

Ranking-Based Siamese Visual Tracking

Tian¹,

Li²

2022

Preprint

View full text Add to dashboard Cite

Current Siamese-based trackers mainly formulate the visual tracking into two independent subtasks, including classification and localization. They learn the classification subnetwork by processing each sample separately and neglect the relationship among positive and negative samples. Moreover, such tracking paradigm takes only the classification confidence of proposals for the final prediction, which may yield the misalignment between classification and localization. To resolve these issues, this paper proposes a ranking-based optimization algorithm to explore the relationship among different proposals. To this end, we introduce two ranking losses, including the classification one and the IoU-guided one, as optimization constraints. The classification ranking loss can ensure that positive samples rank higher than hard negative ones, i.e., distractors, so that the trackers can select the foreground samples successfully without being fooled by the distractors. The IoUguided ranking loss aims to align classification confidence scores with the Intersection over Union(IoU) of the corresponding localization prediction for positive samples, enabling the well-localized prediction to be represented by high classification confidence. Specifically, the proposed two ranking losses are compatible with most Siamese trackers and incur no additional computation for inference. Extensive experiments on seven tracking benchmarks, including OTB100, UAV123, TC128, VOT2016, NFS30, GOT-10k and LaSOT, demonstrate the effectiveness of the proposed ranking-based optimization algorithm. The code and raw results are available at https://github.com/ sansanfree/RBO.

show abstract

Section: Comparison With State-of-the-art Trackersmentioning

confidence: 87%

Ranking-Based Siamese Visual Tracking

Tian¹,

Li²

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…However, the offline target matching with a shallow correlation structure [2] lacks of discriminative power towards distractors. Then, the dedicated modifications rise, including attention mechanism [15,43,56], online module [61,63], cascaded frameworks [7,14,41], update mechanism [57] and target-aware model fine-tuning [24,40]. Despite the improvements, most of them bring much complexity to the Siamese tracking pipeline.…”

Section: Related Workmentioning

confidence: 99%

Correlation-Aware Deep Tracking

Xie¹,

Wang²,

Wang³

et al. 2022

Preprint

View full text Add to dashboard Cite

Robustness and discrimination power are two fundamental requirements in visual object tracking. In most tracking paradigms, we find that the features extracted by the popular Siamese-like networks cannot fully discriminatively model the tracked targets and distractor objects, hindering them from simultaneously meeting these two requirements. While most methods focus on designing robust correlation operations, we propose a novel target-dependent feature network inspired by the self-/cross-attention scheme. In contrast to the Siamese-like feature extraction, our network deeply embeds cross-image feature correlation in multiple layers of the feature network. By extensively matching the features of the two images through multiple layers, it is able to suppress non-target features, resulting in instancevarying feature extraction. The output features of the search image can be directly used for predicting target locations without extra correlation step. Moreover, our model can be flexibly pre-trained on abundant unpaired images, leading to notably faster convergence than the existing methods. Extensive experiments show our method achieves the stateof-the-art results while running at real-time. Our feature networks also can be applied to existing tracking pipelines seamlessly to raise the tracking performance. Code will be available.

show abstract

“…In row 2, another bus similar to the target is an interference for the tracking methods, which reduces the success rate. In row 3, during occlusion, the appearance models, including feature extractor and observation model, keep updated with the unreliable image region in most of the tracking methods [4,6,9,16,21,31], and finally cause the model drift. Taking the advantage of the M-model, the proposed M-CoTransT utilises the reliable historical motion state to estimate the more precise search region for the reappear targets after occlusion.…”

Section: Attributes Evaluationmentioning

confidence: 99%

“…Regarding occlusion, recent tracking methods [7,[9][10][11]] develop a motion model through the normal Markov process efficiently by using the motion state of a target in the last frame to predict the search region. Nevertheless, when occlusion occurs, the motion state in the last frame is unreliable, and the target reappears after occlusion is not in the search region.…”

Section: Introductionmentioning

confidence: 99%

M‐CoTransT: Adaptive spatial continuity in visual tracking

Fan

Zhang

Ming

2022

IET Computer Vision

View full text Add to dashboard Cite

Visual tracking is an important area in computer vision. Based on the Siamese network, current tracking methods employ the self-attention block in convolutional networks to extract semantic features containing the image structure information of an object. However, spatial continuity is a point of contradiction between two seemingly unrelated challenges, that is, occlusion and similar distractor, in tracking methods. At the same time, it is a spatially discontinuous task to locate a target reappearing after occlusion accurately. The prediction of bounding boxes should be constrained by spatial continuity to prevent them from jumping into similar distractors. This study proposes a novel tracking method for introducing spatial continuity in visual tracking called M-CoTransT; the novel tracking method is developed through the confidence-based adaptive Markov motion model (M-model) and a novel correlation-based feature fusion network (CoTransT). In particular, the M-model provides confidence for the nodes of the Markov motion model to estimate the motion state continuity. It also predicts a more accurate search region for CoTransT, which then adds a cross-correlation branch into the self-attention tracking network to enhance the continuity of target appearance in the feature space. Extensive experiments on five challenging datasets (LaSOT, GOT-10k, TrackingNet, OTB-2015 and UAV123) demonstrated the effectiveness of the proposed M-CoTransT in visual tracking.

show abstract

Learning to Filter: Siamese Relation Network for Robust Tracking

Cited by 114 publications

References 42 publications

Ranking-Based Siamese Visual Tracking

Ranking-Based Siamese Visual Tracking

Correlation-Aware Deep Tracking

M‐CoTransT: Adaptive spatial continuity in visual tracking

Contact Info

Product

Resources

About