Visual tracking using Siamese convolutional neural network with region proposal and domain specific updating

Zhang, Han; Ni, Weijian; Yan, Weidong; Jian, Wu; Bu, Hui; Xiang, Deliang

doi:10.1016/j.neucom.2017.11.050

Cited by 26 publications

(14 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, a weighting strategy is incorporated into the Siamese network to make the tracker more versatile in dealing with large object appearance variations. It is worth noting that our proposed method bears some similarity to a recent paper [28] as they incorporate pretrained Siamese network into sequence-specific updating. However, our approach differs from [28] in several respects.…”

Section: Generative Cnn-based Trackersmentioning

confidence: 84%

“…It is worth noting that our proposed method bears some similarity to a recent paper [28] as they incorporate pretrained Siamese network into sequence-specific updating. However, our approach differs from [28] in several respects. First, Our proposed Siamese network is based on the SINT [15] architecture that integrates hierarchical features in order to consider both semantic and spatial details.…”

Section: Generative Cnn-based Trackersmentioning

confidence: 84%

“…To tackle these challenges, we pursue two procedures: 1) Unlike previous Siamese-based trackers [7,15] relying solely on the initial ground truth to find the target, in this work the best previous candidates are stored in a buffer and the new candidates are then assessed with the stored elements. To impede entering the occluded candidates into the buffer, buffer updating is restricted to a predefined condition; 2) Recently, much attention has been given to the use of sequence-specific information for fine-tuning a pretrained CNN [13,17,21,22,28,33]. Integrating online finetuning strategy into a pretrained network can make the tracker adaptive to the large appearance changes of the target and thus enhance the tracker robustness.…”

Section: Baseline Siamese Networkmentioning

confidence: 99%

See 2 more Smart Citations

Visual object tracking based on adaptive Siamese and motion estimation network

Kashiani

Shokouhi

2019

Image and Vision Computing

View full text Add to dashboard Cite

Recently, convolutional neural network (CNN) has attracted much attention in different areas of computer vision, due to its powerful abstract feature representation. Visual object tracking is one of the interesting and important areas in computer vision that achieves remarkable improvements in recent years. In this work, we aim to improve both the motion and observation models in visual object tracking by leveraging representation power of CNNs. To this end, a motion estimation network (named MEN) is utilized to seek the most likely locations of the target and prepare a further clue in addition to the previous target position.Hence the motion estimation would be enhanced by generating a small number of candidates near two plausible positions. The generated candidates are then fed into a trained Siamese network to detect the most probable candidate.Each candidate is compared to an adaptable buffer, which is updated under a predefined condition. To take into account the target appearance changes, a weighting CNN (called WCNN) adaptively assigns weights to the final similarity scores of the Siamese network using sequence-specific information. Evaluation results on well-known benchmark datasets (OTB100, OTB50 and OTB2013) prove that the proposed tracker outperforms the state-of-the-art competitors.

show abstract

Section: Generative Cnn-based Trackersmentioning

confidence: 84%

Section: Generative Cnn-based Trackersmentioning

confidence: 84%

Section: Baseline Siamese Networkmentioning

confidence: 99%

See 1 more Smart Citation

Visual object tracking based on adaptive Siamese and motion estimation network

Kashiani

Shokouhi

2019

Image and Vision Computing

View full text Add to dashboard Cite

show abstract

“…The excellent performance of SiamFC has attracted wide attention in target tracking and its follow-up work [15], [16], [22], [26], [27] . SiamRPN improved the speed and accuracy of tracking algorithm by using the candidate sub-regions recommended by RPN [15].…”

Section: B Pedestrian Tracking Based On Siamese Networkmentioning

confidence: 99%

Infrared Multi-Pedestrian Tracking in Vertical View via Siamese Convolution Network

et al. 2019

View full text Add to dashboard Cite

Target tracking has become one of the research hotspots in the field of computer vision in recent years. In this paper, a new intelligent algorithm of infrared multi-pedestrian tracking in vertical view is proposed. In the algorithm, the pedestrians in the infrared image can be quickly detected and located with the method of the Faster Regions with CNN features (RCNN) and then are tracked with the improved Siamese network. The tracking method based on Siamese network transforms the tracking problem into a similarity verification problem and evaluates the similarity score between new frame feature and target frame feature by convolution network. The candidate region with the highest score is considered as the current position of the target. In this paper, the Siamese network is combined with Faster RCNN for multi-pedestrian tracking. In addition, the tracking results of adjacent frames are introduced into the similarity evaluation of current frames to improve the tracking accuracy when the pedestrian posture changes. The experimental results show that the algorithm has good robustness and tracking result and achieves competitive performance. INDEX TERMS Computer vision, Siamese network, infrared detection, pedestrian tracking, convolution network.

show abstract

“…The SiamRPN++ [ 9 ] introduces a ResNet-driven Siamese tracker, which makes layer-wise and depth-wise aggregations explicit when modeling network architecture, which can improve the accuracy and reduce the model size at the same time. The DCFNet [ 10 ] further combines the Siamese network with region proposal networks, and performs the domain specific updating to achieve a light-weight network in end-to-end learning.…”

Section: Introductionmentioning

confidence: 99%

Proposal-Based Visual Tracking Using Spatial Cascaded Transformed Region Proposal Network

Zhang

Luo

Fan

2020

Sensors

View full text Add to dashboard Cite

Region proposal network (RPN) based trackers employ the classification and regression block to generate the proposals, the proposal that contains the highest similarity score is formulated to be the groundtruth candidate of next frame. However, region proposal network based trackers cannot make the best of the features from different convolutional layers, and the original loss function cannot alleviate the data imbalance issue of the training procedure. We propose the Spatial Cascaded Transformed RPN to combine the RPN and STN (spatial transformer network) together, in order to successfully obtain the proposals of high quality, which can simultaneously improves the robustness. The STN can transfer the spatial transformed features though different stages, which extends the spatial representation capability of such networks handling complex scenarios such as scale variation and affine transformation. We break the restriction though an easy samples penalization loss (shrinkage loss) instead of smooth L1 function. Moreover, we perform the multi-cue proposals re-ranking to guarantee the accuracy of the proposed tracker. We extensively prove the effectiveness of our proposed method on the ablation studies of the tracking datasets, which include OTB-2015 (Object Tracking Benchmark 2015), VOT-2018 (Visual Object Tracking 2018), LaSOT (Large Scale Single Object Tracking), TrackingNet (A Large-Scale Dataset and Benchmark for Object Tracking in the Wild) and UAV123 (UAV Tracking Dataset).

show abstract

Visual tracking using Siamese convolutional neural network with region proposal and domain specific updating

Cited by 26 publications

References 7 publications

Visual object tracking based on adaptive Siamese and motion estimation network

Visual object tracking based on adaptive Siamese and motion estimation network

Infrared Multi-Pedestrian Tracking in Vertical View via Siamese Convolution Network

Proposal-Based Visual Tracking Using Spatial Cascaded Transformed Region Proposal Network

Contact Info

Product

Resources

About