Filters








80 Hits in 3.8 sec

Robust Object Tracking via Hierarchical Estimation in SiamFC Network

Yueyang Gu, Yu Qiao, Kuan Xu, Jie Yang, Jiaxuan Xie, Lu Wang, Mingna Liu
2019 Australian Journal of Intelligent Information Processing Systems  
Recently, SiamFC network has been investigated into object tracking community.  ...  We firstly construct hierarchical SiamFC networks using pre-trained VGGNet, ResNet and DenseNet. While tracking, hierarchical SiamFC network can infer target position via different conv-features.  ...  Conclusion In this paper, we propose a hierarchical estimation framework for robust object tracking. We construct hierarchical SiamFC networks using VGGNet, ResNet and DenseNet.  ... 
dblp:journals/ajiips/GuQXYXWL19 fatcat:q5c4rfefizbhfpipyazb22vbqe

RPT: Learning Point Set Representation for Siamese Visual Tracking [article]

Ziang Ma, Linyuan Wang, Haitao Zhang, Wei Lu, Jun Yin
2020 arXiv   pre-print
While remarkable progress has been made in robust visual tracking, accurate target state estimation still remains a highly challenging problem.  ...  In this paper, we argue that this issue is closely related to the prevalent bounding box representation, which provides only a coarse spatial extent of object.  ...  Target Estimation with Point Set Representation In contrast to object detection, tracking target can be an arbitrary object with unknown class.  ... 
arXiv:2008.03467v2 fatcat:btzeyxmztjazrmou3o4utg7fyu

PatchNet – Short-range Template Matching for Efficient Video Processing [article]

Huizi Mao, Sibo Zhu, Song Han, William J. Dally
2021 arXiv   pre-print
We demonstrate its application on two tasks, video object detection and visual object tracking.  ...  We propose PatchNet, an efficient convolutional neural network to match objects in adjacent video frames. It learns the patchwise correlation features instead of pixel features.  ...  Visual object tracking We select SiamFC and SiamRPN as our baseline models. SiamFC is the predecessor of recent advances of Siamese models for VOT.  ... 
arXiv:2103.07371v1 fatcat:ejzywqowrjexfcrpibje7vxsvm

Robust visual tracking via an improved background aware correlation filter

Xiaoxiao Sheng, Yungang Liu, Huijun Liang, Fengzhong Li, Yongchao Man
2019 IEEE Access  
Recently, there emerge many excellent algorithms in the field of visual object tracking.  ...  To overcome the imperfections, an improved BACF method of robust visual object tracking is proposed to achieve the location of targets with higher accuracy in complex scenarios allowing scale variation  ...  It exploits hierarchical convolutional features to improve tracking accuracy and robustness.  ... 
doi:10.1109/access.2019.2900666 fatcat:ux5kgtrrnrdzvd2kbikmfs3yh4

TDIOT: Target-driven Inference for Deep Video Object Tracking [article]

Filiz Gurkan, Llukman Cerkezi, Ozgun Cirakman, Bilge Gunsel
2021 arXiv   pre-print
Moreover, in order to improve robustness to scale changes, we introduce a scale adaptive region proposal network that enables to search the target at an adaptively enlarged spatial neighborhood specified  ...  In order to tackle tracking discontinuities, we incorporate a local search and matching module into the inference head layer that exploits SiamFC for short term tracking.  ...  their class labels and final locations are estimated via classification and regression layers.  ... 
arXiv:2103.11017v2 fatcat:27zliegj6fd37dnkylcikzde5q

Online Decision Based Visual Tracking via Reinforcement Learning

Ke Song, Wei Zhang, Ran Song, Yibin Li
2020 Neural Information Processing Systems  
are adept in.  ...  Unlike previous fusion-based methods, we propose a novel ensemble framework, named DTNet, with an online decision mechanism for visual tracking based on hierarchical reinforcement learning.  ...  Introduction As a fundamental task in computer vision, visual tracking aims to estimate the trajectory of a specified object in a sequence of images.  ... 
dblp:conf/nips/Song0SL20 fatcat:qw7gs5phtnacpbvs2udlid4d6y

Combined Kalman Filter and Multifeature Fusion Siamese Network for Real-Time Visual Tracking

Lijun Zhou, Jianlin Zhang
2019 Sensors  
This also enables our tracker to stably track fast moving targets.The introduction of the Kalman filter compensates for the shortcomings that SiamFC can only track offline, and the tracking network has  ...  SiamFC has a simple network structure and can be pretrained offline on a large data set, so it has attracted the attention of many researchers. It has no online learning process at all.  ...  object tracking by estimating the state of the object.  ... 
doi:10.3390/s19092201 fatcat:dkgkgrh6wzayfdifde3l7kpvhe

Discriminative Siamese Tracker Based on Multi-Channel-Aware and Adaptive Hierarchical Deep Features

Huanlong Zhang, Rui Duan, Anping Zheng, Jie Zhang, Linwei Li, Fengxian Wang
2021 Symmetry  
The Siamese network used in this paper is a two-input branch symmetric neural network with two input branches, and they share the same weights, which are widely used in the field of target tracking.  ...  Finally, the proposed two modules are integrated into the Siamese framework for target tracking.  ...  We integrate the two methods with a Siamese network for object tracking and evaluate the proposed method on some benchmarks.  ... 
doi:10.3390/sym13122329 fatcat:nkbguaakdvfxtoydtm4lifepai

Visual Object Tracking based on Adaptive Siamese and Motion Estimation Network [article]

Hossein Kashiani, Shahriar B. Shokouhi
2018 arXiv   pre-print
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.  ...  Although training a CNN with multiple pooling operations makes the network more robust in dealing with object deformation Algorithm 1 1 Online tracking algorithm Input: Pretrained F-MEN, Pretrained Siamese  ... 
arXiv:1810.00119v1 fatcat:dlkc4ijrbbcxrmcqvz5oth2ruu

An Anchor-Free Siamese Network with Multi-Template Update for Object Tracking

Tongtong Yuan, Wenzhu Yang, Qian Li, Yuxia Wang
2021 Electronics  
We use the anchor-free network to implement object tracking in a per-pixel manner, which computes the object category and bounding boxes directly.  ...  Inspired by the Siamese network and anchor-free idea, an anchor-free Siamese network (AFSN) with multi-template updates for object tracking is proposed.  ...  Acknowledgments: We thank the machine vision lab in Hebei University for the equipment and other help it offered. Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/electronics10091067 fatcat:s7glzec5kjbjtlsockynm3a7ci

Distractor-aware Siamese Networks for Visual Object Tracking [article]

Zheng Zhu, Qiang Wang, Bo Li, Wei Wu, Junjie Yan, Weiming Hu
2018 arXiv   pre-print
Recently, Siamese networks have drawn great attention in visual tracking community because of their balanced accuracy and speed.  ...  In this paper, we focus on learning distractor-aware Siamese networks for accurate and long-term tracking. To this end, features used in traditional Siamese trackers are analyzed at first.  ...  .: Accurate scale estimation for robust visual tracking. In: Proceedings of the British Machine Vision Conference. pp. 65.1-65.11 (2014) 5.  ... 
arXiv:1808.06048v1 fatcat:hwtscbonajh2boqswdvra3b4xm

Enhancement: SiamFC Tracker Algorithm Performance Based on Convolutional Hyperparameters Optimization and Low Pass Filter

Rogeany Kanza, Yu Zhao, Zhilin Huang, Chenyu Huang, Zhuoming Li
2022 Mathematics  
We evaluate our methods on the Visual Object Tracking (VOT) Challenge 2016 dataset, and they both outperform the original SiamFC tracker performance and many other top performers.  ...  in IOU compared to the original SiamFC.  ...  Then, the tracking object is estimated by calculating the cosine distance between the candidate target's feature vector and the vector in the feature vector set.  ... 
doi:10.3390/math10091527 fatcat:3ueblirq55a73m6fmk52y6fm2m

UCT: Learning Unified Convolutional Networks for Real-time Visual Tracking [article]

Zheng Zhu, Guan Huang, Wei Zou, Dalong Du, Chang Huang
2017 arXiv   pre-print
Convolutional neural networks (CNN) based tracking approaches have shown favorable performance in recent benchmarks.  ...  The standard UCT and UCT-Lite can track generic objects at 41 FPS and 154 FPS without further optimization, respectively.  ...  The region proposal network (RPN) in Faster R-CNN [20] is a fully convolutional network that simultaneously predicts object bounds and objectness scores at each position.  ... 
arXiv:1711.04661v1 fatcat:vfo73m7iq5hndlzelzob6dotiu

Real Time Visual Tracking using Spatial-Aware Temporal Aggregation Network [article]

Tao Hu, Lichao Huang, Xianming Liu, Han Shen
2019 arXiv   pre-print
In addition, we also use a feature pyramid structure to handle motion estimation at different scales, and address the different demands on feature granularity between tracking losses and deformation offset  ...  More powerful feature representations derived from deep neural networks benefit visual tracking algorithms widely.  ...  In this work, we present a novel spatial-aware temporal feature aggregation network for visual object tracking as shown in Figure 2 .  ... 
arXiv:1908.00692v1 fatcat:acwywix2jbgkbicu6ysgordjfu

UCT: Learning Unified Convolutional Networks for Real-Time Visual Tracking

Zheng Zhu, Guan Huang, Wei Zou, Dalong Du, Chang Huang
2017 2017 IEEE International Conference on Computer Vision Workshops (ICCVW)  
Convolutional neural networks (CNN) based tracking approaches have shown favorable performance in recent benchmarks.  ...  The standard UCT and UCT-Lite can track generic objects at 41 FPS and 154 FPS without further optimization, respectively.  ...  The region proposal network (RPN) in Faster R-CNN [20] is a fully convolutional network that simultaneously predicts object bounds and objectness scores at each position.  ... 
doi:10.1109/iccvw.2017.231 dblp:conf/iccvw/ZhuHZDH17 fatcat:tivrh5ke2jg55dxc65hx4oilva
« Previous Showing results 1 — 15 out of 80 results