Filters








1,843 Hits in 3.5 sec

A Memory Model Based on the Siamese Network for Long-Term Tracking [chapter]

Hankyeol Lee, Seokeon Choi, Changick Kim
2019 Lecture Notes in Computer Science  
We propose a novel memory model using deep convolutional features for long-term tracking to handle the challenging issues, including visual deformation or target disappearance.  ...  Our memory model is separated into short-and long-term stores inspired by Atkinson-Shiffrin Memory Model (ASMM).  ...  Conclusion In this paper, we have proposed a memory model based on the Siamese network for long-term tracking (MMLT).  ... 
doi:10.1007/978-3-030-11009-3_5 fatcat:pft36ndzcvfyxaab7ugecgamse

DMV: Visual Object Tracking via Part-level Dense Memory and Voting-based Retrieval [article]

Gunhee Nam, Seoung Wug Oh, Joon-Young Lee, Seon Joo Kim
2020 arXiv   pre-print
We propose a novel memory-based tracker via part-level dense memory and voting-based retrieval, called DMV.  ...  We also propose a novel voting mechanism for the memory reading to filter out unreliable information in the memory.  ...  [25] proposed memory trackers for long term tracking by interpolating the multiple estimated templates stored in a short term and a long term memory according to the number of successive frames with  ... 
arXiv:2003.09171v1 fatcat:ulnrrajatvcp5gzt6lk4alzjwy

Tracking Holistic Object Representations [article]

Axel Sauer, Elie Aljalbout, Sami Haddadin
2019 arXiv   pre-print
We propose a framework that is designed to be used on top of previous trackers without any need for further training of the siamese network.  ...  Recent advances in visual tracking are based on siamese feature extractors and template matching.  ...  We gratefully acknowledge the general support of Microsoft Germany and the Alfried Krupp von Bohlen und Halbach Foundation.  ... 
arXiv:1907.12920v2 fatcat:32a7xlkd55c2ziipfdhgs67f7i

An Anti-UAV Long-Term Tracking Method with Hybrid Attention Mechanism and Hierarchical Discriminator

Feng Cheng, Zhibo Liang, Gaoliang Peng, Shaohui Liu, Sijue Li, Mengyu Ji
2022 Sensors  
At the same time, the attention-based Siamese network fuses multilevel features for accurately tracking the target.  ...  In this article, a novel long-term tracking architecture composed of a Siamese network and re-detection (SiamAD) is proposed to efficiently locate UAV targets in diverse surroundings.  ...  Previously, [16, 17] proposed a detection module using a global search strategy based on a Siamese network. The above long-term tracking methods are more robust in practical applications.  ... 
doi:10.3390/s22103701 pmid:35632110 fatcat:3qvfzk7vmjdtjhrg3bs3rvfmci

Distractor-Aware Siamese Networks for Visual Object Tracking [chapter]

Zheng Zhu, Qiang Wang, Bo Li, Wei Wu, Junjie Yan, Weiming Hu
2018 Lecture Notes in Computer Science  
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.  ...  In addition, we extend the proposed approach for long-term tracking by introducing a simple yet effective local-to-global search region strategy.  ...  Related Work Siamese Networks based Tracking. Siamese trackers follow a tracking by similarity comparison strategy.  ... 
doi:10.1007/978-3-030-01240-3_7 fatcat:g2fjqyj7uzg5li2afim5kegy64

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

Hossein Kashiani, Shahriar B. Shokouhi
2018 arXiv   pre-print
The generated candidates are then fed into a trained Siamese network to detect the most probable candidate.  ...  In this work, we aim to improve both the motion and observation models in visual object tracking by leveraging representation power of CNNs.  ...  Based on the observation model, CNN-based visual object tracking is approximately divided into discriminative and generative approaches [11] .  ... 
arXiv:1810.00119v1 fatcat:dlkc4ijrbbcxrmcqvz5oth2ruu

SiamRPN++: Evolution of Siamese Visual Tracking with Very Deep Networks [article]

Bo Li, Wei Wu, Qiang Wang, Fangyi Zhang, Junliang Xing, Junjie Yan
2018 arXiv   pre-print
Siamese network based trackers formulate tracking as convolutional feature cross-correlation between target template and searching region.  ...  Our model will be released to facilitate further studies based on this problem.  ...  Analysis on Siamese Networks for Tracking The Siamese network based tracking algorithms [40, 1] formulate visual tracking as a cross-correlation problem and learn a tracking similarity map from deep  ... 
arXiv:1812.11703v1 fatcat:lugxugdp5rhixpz3peuem5cqsa

Multi-Object Tracking with Deep Learning Ensemble for Unmanned Aerial System Applications

Wanlin Xie, Jaime Ide, Daniel Izadi, Sean Banger, Thayne Walker, Ryan Ceresani, Dylan Spagnuolo, Christopher Guagliano, Henry Diaz, Jason Twedt
2021 arXiv   pre-print
For the visual scoring, we experiment with different similarity measures to calculate distance based on entity appearances, including a convolutional neural network (CNN) encoder, pre-trained using Siamese  ...  In this work, we present a robust object tracking architecture aimed to accommodate for the noise in real-time situations.  ...  We thank Andrew Walsh and Kenneth Rapko Jr. for their valuable feedback on benchmarking techniques in object tracking.  ... 
arXiv:2110.02044v1 fatcat:ljcoxyurpfba7ncwn75jfrtff4

Discriminative and Robust Online Learning for Siamese Visual Tracking [article]

Jinghao Zhou, Peng Wang, Haoyang Sun
2019 arXiv   pre-print
Beyond that, our model based on SiamRPN++ obtains the best results over six popular tracking benchmarks and can operate beyond real-time.  ...  The problem of visual object tracking has traditionally been handled by variant tracking paradigms, either learning a model of the object's appearance exclusively online or matching the object with the  ...  The template for this branch is denoted as z s , representing the template in short-term memory as opposed to retaining the first frame for long-term memory.  ... 
arXiv:1909.02959v2 fatcat:pmc5vg3odnghnjotg4wzvfjcmy

Online Multiple Object Tracking Using Rule Distillated Siamese Random Forest

Jimi Lee, Sang Won Kim, Byoung Chul Ko
2020 IEEE Access  
Although a long-term appearance model using features from a DNN [3] , DeepMatching [4] , and a quadruplet convolutional neural network (CNN) [5] have been proposed, offline tracking is unsuitable for  ...  The external memory explicitly stores previous inputs of each trajectory in a time window, whereas the internal memory learns to summarize the long-term tracking history and associate detections by processing  ... 
doi:10.1109/access.2020.3028770 fatcat:3vwgdbui6ff7xgprxgexbby4di

Deep Siamese Networks with Bayesian non-Parametrics for Video Object Tracking [article]

Anthony D. Rhodes, Manan Goel
2018 arXiv   pre-print
Using Gaussian Process priors, we model a dynamic objective function representing the location of a tracked object in each frame.  ...  for efficient object tracking in video.  ...  Siamese Networks We adopt the Siamese network-based approach for oneshot image recognition from [15] to learn a generic, deep similarity function for object tracking.  ... 
arXiv:1811.07386v1 fatcat:ysaillmbl5dfzbjunbmvcabww4

Discriminative and Robust Online Learning for Siamese Visual Tracking

Jinghao Zhou, Peng Wang, Haoyang Sun
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Beyond that, our model based on SiamRPN++ obtains the best results over six popular tracking benchmarks and can operate beyond real-time.  ...  The problem of visual object tracking has traditionally been handled by variant tracking paradigms, either learning a model of the object's appearance exclusively online or matching the object with the  ...  The template for this branch is denoted as z s , representing the template in short-term memory as opposed to retaining the first frame for long-term memory.  ... 
doi:10.1609/aaai.v34i07.7002 fatcat:vrx3qxk5qrcbtdpfvlbp5mmelm

Learning Dynamic Compact Memory Embedding for Deformable Visual Object Tracking [article]

Pengfei Zhu, Hongtao Yu, Kaihua Zhang, Yu Wang, Shuai Zhao, Lei Wang, Tianzhu Zhang, Qinghua Hu
2021 arXiv   pre-print
To this end, we propose a dynamic compact memory embedding to enhance the discrimination of the segmentation-based deformable visual tracking method.  ...  Recently, template-based trackers have become the leading tracking algorithms with promising performance in terms of efficiency and accuracy.  ...  RELATED WORK A. Visual Object Tracking Siamese Network-based Tracking.  ... 
arXiv:2111.11625v1 fatcat:qlf4e4q5yvglxannl3xqzgygaq

The Sixth Visual Object Tracking VOT2018 Challenge Results [chapter]

Matej Kristan, Aleš Leonardis, Jiří Matas, Michael Felsberg, Roman Pflugfelder, Luka Čehovin Zajc, Tomáš Vojír̃, Goutam Bhat, Alan Lukežič, Abdelrahman Eldesokey, Gustavo Fernández, Álvaro García-Martín (+143 others)
2019 Lecture Notes in Computer Science  
A new dataset has been compiled and a performance evaluation methodology that focuses on long-term tracking capabilities has been adopted.  ...  The new subchallenge focuses on long-term tracking properties, namely coping with target disappearance and reappearance.  ...  B.1 A Memory Model based on the Siamese Network for Long-term Tracking (MMLT) H. Lee, S. Choi, C.  ... 
doi:10.1007/978-3-030-11009-3_1 fatcat:wbfi2kp3rze33cl2o56eauhooq

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

Madhu Kiran, Vivek Tiwari, Le Thanh Nguyen-Meidine, Eric Granger
2019 arXiv   pre-print
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 detector-tracker interactions for trackers from the Siamese-FC family.  ...  Siamese Fully-Convolutional (SiamFC) tracker [1] uses an AlexNet based Siamese network for feature extraction.  ... 
arXiv:1910.14552v1 fatcat:vj7b6j4dkvhrrm4gib3giug4by
« Previous Showing results 1 — 15 out of 1,843 results