104,450 Hits in 4.4 sec

Beyond Background-Aware Correlation Filters: Adaptive Context Modeling by Hand-Crafted and Deep RGB Features for Visual Tracking [article]

Seyed Mojtaba Marvasti-Zadeh, Hossein Ghanei-Yakhdan, Shohreh Kasaei
2021 arXiv   pre-print
In recent years, the background-aware correlation filters have achie-ved a lot of research interest in the visual target tracking.  ...  Comprehensive experimental results demonstrate that the proposed adaptive method clearly outperforms the accuracy and robustness of visual target tracking compared to the state-of-the-art methods on the  ...  Conclusion An adaptive context-aware visual tracking method which exploits both handcrafted and deep RGB feature maps for context modeling was proposed.  ... 
arXiv:2004.02932v2 fatcat:dbjgzsequvcpxgq2qlilv6rmbi

Overview of contextual tracking approaches in information fusion

Erik Blasch, Jesus Garcia Herrero, Lauro Snidaro, James Llinas, Guna Seetharaman, Kannappan Palaniappan, Matthew F. Pellechia, Richard J. Sorensen, Kannappan Palaniappan
2013 Geospatial InfoFusion III  
for target tracking and identification.  ...  knowledge target tracking algorithm construction, implementation, and application.  ...  Similarly, methods of mathematical performance analysis were developed for maneuvering target pose [39] , maneuvering target state estimation [40] and tracking robustness for spatial-temporal context  ... 
doi:10.1117/12.2016312 fatcat:l24tj2zkqzgc7cvdsssuupipw4

Learning Cascaded Context-Aware Framework for Robust Visual Tracking

Ding Ma, Xiangqian Wu
2019 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)  
Context information on each corner of the whole image is useful for visual tracking. However, some trackers may not be able to model such information, this will result in suboptimal performance.  ...  Depending on the output of these two networks-the final context-aware map, we can generate the bounding box of the target flexibly.  ...  By taking advantage of the cascaded structure, we propose a cascaded context-aware framework for visual tracking.  ... 
doi:10.1109/iccvw.2019.00010 dblp:conf/iccvw/Ma019 fatcat:ogpqub5ptjbzpkjjdxpbtgjuoe

Object Tracking in Unmanned Aerial Vehicle Videos via Multifeature Discrimination and Instance-Aware Attention Network

Shiyu Zhang, Li Zhuo, Hui Zhang, Jiafeng Li
2020 Remote Sensing  
In the bounding-box estimation stage, an instance-aware intersection over union (IoU)-Net is integrated together with an instance-aware attention network to estimate the target size based on the bounding-box  ...  In this paper, we propose a novel two-stage UAV tracking framework, which includes a target detection stage based on multifeature discrimination and a bounding-box estimation stage based on the instance-aware  ...  For the estimation task, it modifies IoU-Net [14] to perform target-specific prediction to estimate the target bounding-box, which is suitable for tracking tasks.  ... 
doi:10.3390/rs12162646 fatcat:5jdfgen6fbaiplccvqkzowsgaa

Context-Aware and Occlusion Handling Mechanism for Online Visual Object Tracking

Khizer Mehmood, Abdul Jalil, Ahmad Ali, Baber Khan, Maria Murad, Wasim Ullah Khan, Yigang He
2020 Electronics  
In the present study, an adaptive Spatio-temporal context (STC)-based algorithm for online tracking is proposed by combining the context-aware formulation, Kalman filter, and adaptive model learning rate  ...  The state of an object in the tracking process depends on the maximum value of the response map between consecutive frames.  ...  Related Work With recent advancement in visual tracking, various competitive methods have been proposed for target tracking. Zhang et al.  ... 
doi:10.3390/electronics10010043 fatcat:5g2ybycivbbdjdzvijjgfiis6a

Trajectory Smoothing Constraint and Hard Negative Mining for Distractor-aware Regression Tracking

Weichun Liu, Xiaoan Tang, Xiaoyuan Ren
2019 IEEE Access  
In this paper, we propose to achieve distractor-aware regression tracking with trajectory smoothing constraint and hard negative mining.  ...  Recently, convolutional regression networks have drawn great attention in the tracking community.  ...  Output: Estimated target state X t = (x t , y t , s t ) in each frame.  ... 
doi:10.1109/access.2019.2921562 fatcat:vk6u74aunrc3zahc7nhwdwg5oe

Visual Multiple Target Tracking From a Descending Aerial Platform

Parker C. Lusk, Randal W. Beard
2018 2018 Annual American Control Conference (ACC)  
The recently developed Recursive-RANSAC algorithm uses the visual measurements to estimate targets in clutter.  ...  A real-time visual multiple target tracker is demonstrated onboard a descending multirotor. Measurements of moving ground targets are generated using the Kanade-Lucas-Tomasi (KLT) tracking method.  ...  ACKNOWLEDGMENT The authors thank the NASA Langley Systems Analysis & Concepts Directorate (SACD) and the NASA UAS Traffic Management (UTM) project for funding this work and providing flight test support  ... 
doi:10.23919/acc.2018.8431915 dblp:conf/amcc/LuskB18 fatcat:cequeuftbbfmfetuq3fbenzgua

Target-Aware Deep Tracking

Xin Li, Chao Ma, Baoyuan Wu, Zhenyu He, Ming-Hsuan Yang
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
The target-aware features are integrated with a Siamese matching network for visual tracking.  ...  The key issue is that in visual tracking the targets of interest can be arbitrary object class with arbitrary forms.  ...  Target-Aware Features In this section, we present how to learn target-aware features for visual tracking.  ... 
doi:10.1109/cvpr.2019.00146 dblp:conf/cvpr/Li0WH019 fatcat:hjod2ekhzbfjdonfhfxi6iekha

Learning a unified tracking-and-detection framework with distractor-aware constraint for visual object tracking

Yang Fang, Seunghyun Ko, Geun-Sik Jo
2020 The Journal of Engineering  
In this study, the authors propose a unified tracking-and-detection framework with distractor-aware (UTDF-DA), which involves both context learning and target reidentification with a target-aware detector  ...  Then a single-shot multibox detector-based target-aware detector is trained by domain-specific meta-training approach for deep detection features and hard-negative samples generation.  ...  We first present distractor-aware constraint formulation with state-of-the-art CCF for context knowledge learning.  ... 
doi:10.1049/joe.2019.1205 fatcat:5mklk7qzjzhdbacez7vtioyyce

Demonstrator: V2V Communications in Automotive Multi-Sensor Multi-Target Tracking

Matthias Rockl, Thomas Strang, Matthias Kranz
2008 2008 IEEE 68th Vehicular Technology Conference  
Therefore, we introduce multi-target tracking including Vehicle-2-Vehicle communications as a complementing sensor for future driver assistance systems.  ...  Today's automotive sensor systems for in-vehicle based target tracking, i.e. radar, lidar, camera, are limited to a field of view which is restricted by distance, angle and line-of-sight.  ...  In order to enable the tracking of multiple targets we use a state model that is separated in n partitions including the state estimation for n target vehicles.  ... 
doi:10.1109/vetecf.2008.463 dblp:conf/vtc/RocklSK08a fatcat:ig4cl6igazbhxmurstw26zwpiu

User awareness of tracking uncertainties in AR navigation scenarios

Frieder Pankratz, Andreas Dippon, Tayfur Coskun, Gudrun Klinker
2013 2013 IEEE International Symposium on Mixed and Augmented Reality (ISMAR)  
(c, d) ABSTRACT Current Augmented Reality navigation applications for pedestrians usually do not visualize tracking errors.  ...  To increase the awareness of users about potential imperfections of the tracking at a given time, we alter the visualization of the navigation system.  ...  ACKNOWLEDGEMENTS We thank Manuel Huber and Marcus Tönnis for their contribution. This work was supported by the European CELTIC project CRUMBS CP7-003 and funded by BMWi.  ... 
doi:10.1109/ismar.2013.6671807 dblp:conf/ismar/PankratzDCK13 fatcat:rmszc36jnncexac7g37qmkdq4u

Vehicle Trajectory Estimation Based on Fusion of Visual Motion Features and Deep Learning

Lianen Qu, Matthew N. Dailey
2021 Sensors  
The results show that the algorithm is superior to state-of-the-art sequential state estimation methods such as visual SLAM in performing accurate global localization and trajectory estimation for host  ...  Driver situation awareness is critical for safety. In this paper, we propose a fast, accurate method for obtaining real-time situation awareness using a single type of sensor: monocular cameras.  ...  Visual Track Handling The tracking method integrates the predicted state xt , the detected bounding box a i for each detection i, and the predicted bounding box b j for each tracked target j.  ... 
doi:10.3390/s21237969 pmid:34883970 fatcat:cwpsfhqu2bew5nxmfyp35um3xa

COMET: Context-Aware IoU-Guided Network for Small Object Tracking [article]

Seyed Mojtaba Marvasti-Zadeh, Javad Khaghani, Hossein Ghanei-Yakhdan, Shohreh Kasaei, Li Cheng
2020 arXiv   pre-print
Empirically, COMET outperforms the state-of-the-arts in a range of aerial view datasets that focusing on tracking small objects.  ...  We consider the problem of tracking an unknown small target from aerial videos of medium to high altitudes.  ...  State-of-the-art Comparison For quantitative comparison, the proposed method (COMET) is compared with state-of-the-art visual tracking methods including AutoTrack [24] , ATOM [25] , DiMP-50 [38] , SiamRPN  ... 
arXiv:2006.02597v3 fatcat:m2jmlmdryvbplfv6xkhgm6jxzi

Target-Aware Deep Tracking [article]

Xin Li, Chao Ma, Baoyuan Wu, Zhenyu He, Ming-Hsuan Yang
2019 arXiv   pre-print
The target-aware features are integrated with a Siamese matching network for visual tracking.  ...  The key issue is that in visual tracking the targets of interest can be arbitrary object class with arbitrary forms.  ...  and target objects of arbitrary forms for visual tracking. • We integrate the target-aware features with a Siamese matching network for visual tracking.  ... 
arXiv:1904.01772v1 fatcat:ttqdj7sc4jaxxanbaamsuiv6ye

Hierarchical Spatial-aware Siamese Network for Thermal Infrared Object Tracking [article]

Xin Li, Qiao Liu, Nana Fan, Zhenyu He, Hongzhi Wang
2018 arXiv   pre-print
Finally, we locate the candidate that is most similar to the tracked target.  ...  Next, this pre-trained Siamese network is used to evaluate the similarity between the target template and target candidates.  ...  Algorithm 1 The proposed tracker (HSSNet) 1: Inputs: initial target state x 1 , the learned similarity function f using Eq. 10. 2: Outputs: the estimated target statex t . 3: while t < length(sequence)  ... 
arXiv:1711.09539v2 fatcat:lde6wirsp5ffnodxl5ohg5ouba
« Previous Showing results 1 — 15 out of 104,450 results