6,820 Hits in 4.0 sec

Visual Tracking Using L2 Minimization

Zhijun Pei, Lei Han, S.A. Hamouda, M. Mirzaei, Z. Yu
2016 MATEC Web of Conferences  
Due to the need to solve the L1 norm related minimization problem for many times, this L1 tracker is very computationally demanding.  ...  It may be possible to apply the L2 minimization, instead of L1 minimization, to the visual tracking problems, which has been verified by experiments on challenging sequences in the paper.  ...  and using the L2-norm instead of L1-norm [6] .  ... 
doi:10.1051/matecconf/20166102020 fatcat:w2j5u5bbpjhdviwovjyyyi25bu

Robust Visual Tracking with Improved Collaborative Representation

Chenhua Liu, Jun Kong, Min Jiang, Shengwei Tian
2016 Innovative Computing Information and Control Express Letters, Part B: Applications  
Visual tracking is a challenging problem for the appearance changes caused by extrinsic and intrinsic factors.  ...  In this paper, a robust object tracking algorithm exploiting both collaborative representation and coding residual is proposed within the Bayesian inference framework.  ...  Therefore, our tracker is more robust to occlusion compared with L2-RLS tracker.  ... 
doi:10.24507/icicelb.07.04.789 fatcat:s4kup6oq3nhgvjrxsp4aaufaoq

Video stabilization with L1–L2 optimization

Hui Qu, Li Song
2013 2013 IEEE International Conference on Image Processing  
In the proposed algorithm, we compute smoothed camera paths that are composed of constant, linear and parabolic segments by L1 constraints, meanwhile using the L2 norm of the difference between smoothed  ...  In this paper we present a novel video stabilization algorithm by mixed L1-L2 optimization, aiming at removing unwanted camera movements as well as keeping the original video information to the greatest  ...  be close to constant due to L2 norm.  ... 
doi:10.1109/icip.2013.6738007 dblp:conf/icip/QuS13 fatcat:usup7r6v7fg7jcqkrzaiffj3dm

Online Visual Tracking with Improved Sparse Prototypes

Hongli Yan, Chenhua Liu, Yonghua Shi
2016 Innovative Computing Information and Control Express Letters, Part B: Applications  
Different from the traditional sparse prototypes based tracking method, we employ the sparse representation for the observation instead of the collaborative representation, which can effectively reject  ...  In this paper, a robust tracking method based on improved sparse prototypes is proposed by solving a Bayesian inference problem.  ...  Online Update: The online update of target subspace is important to the changes of object in the process of tracking.  ... 
doi:10.24507/icicelb.07.09.1885 fatcat:m7x2wuspzzd4vfwusbrnuciwpa

Sparse Representation Tracking with Auxiliary Adaptive Appearance Models

Junqiu Wang, Chao Yang, Junqiu Wang, Zhichao Gan, Yasushi Yagi
2017 American Journal of Engineering and Applied Sciences  
The number of target candidates can be very large when the state space is densely sampled. Each minimization takes long time to find the solution.  ...  In this work, we found that l1 minimization can be limited to a few regions with the reasonable probability based on adaptive appearance modeling and background probability estimation.  ...  Author's Contributions Junqiu Wang: He designed the research, implemented the system, did the experiments. He also prepared the initial draft of the manuscript.  ... 
doi:10.3844/ajeassp.2017.51.58 fatcat:x362jr56mndwxedxsd4qimeigm

Predictor-based neural dynamic surface control for distributed formation tracking of multiple marine surface vehicles with improved transient performance

Zhouhua Peng, Dan Wang, Tieshan Li
2016 Science China Information Sciences  
The objective is to achieve a collective tracking with a time-varying trajectory, which can only be accessed by a fraction of follower vehicles.  ...  In this paper, we investigate the distributed formation tracking problem of multiple marine surface vehicles with model uncertainty and time-varying ocean disturbances induced by wind, waves, and ocean  ...  Acknowledgements The authors would like to thank the reviewers for their constructive comments, which have improved the quality of this paper.  ... 
doi:10.1007/s11432-015-5384-9 fatcat:erufja456ne6bcsebxequtirke

Robust iterative learning control design is straightforward for uncertain LTI systems satisfying the robust performance condition

A. Tayebi, M.B. Zaremba
2003 IEEE Transactions on Automatic Control  
ACKNOWLEDGMENT The authors would like to thank the anonymous reviewers for their valuable comments.  ...  In this paper, we prove the exponential convergence of the L2-norm-which is more effective than the -norm-of the tracking error.  ...  1, in the sense of the L 2 -norm; ii) the least upper bound of the L2-norm of the final tracking error is less than the least upper bound of the L 2 -norm of the initial tracking error, i.e., ke1k2 1,  ... 
doi:10.1109/tac.2002.806659 fatcat:uee3gvyrrvb3xerxwl3wdnukzq

Iteratively Reweighted Sparse Coding for Visual Tracking

Hongli Yan, Qibin Lin, Yufeng Wang
2016 Innovative Computing Information and Control Express Letters, Part B: Applications  
Recently the sparse representation has been successfully applied to visual tracking.  ...  In this paper, we model the sparse coding as a sparsity-constrained problem with a weighted matrix in the framework of Bayesian inference, which can be more robust to reject outliers.  ...  [9] introduce square trivial templates in the subspace representation, and employ the L2-norm to regulate the error coefficient.  ... 
doi:10.24507/icicelb.07.12.2619 fatcat:5zstcanqefgq7cglhmushaalfi


Eduardo J. Adam
2020 Zenodo  
The strong nonlinearities together with the constrains of the plant can lead to a non-monotonic convergence of the l2-norm of the error, and still worse, an unstable equilibrium signal e∞(t) can be reached  ...  By numeric simulation this works shows that with the adaptive feedback-based ILC is possible to obtain a better performance in the controlled variable than with the traditional feedback and the feedback  ...  sense of the l2-norm.  ... 
doi:10.5281/zenodo.4312313 fatcat:fhyngdo7ybbf7kteva5jnvwzhi

Splitting Methods for Convex Clustering

Eric C. Chi, Kenneth Lange
2015 Journal of Computational And Graphical Statistics  
the door to potentially novel norms.  ...  In contrast to previously considered algorithms, our ADMM and AMA formulations provide simple and unified frameworks for solving the convex clustering problem under the previously studied norms and open  ...  Applying the projection formula (3.15) to the v l update (3.11) yields the revised update v m+1 l = u m+1 l1 − u m+1 l2 − ν −1 λ m l − P tB [u m+1 l1 − u m+1 l2 − ν −1 λ m l ], for the constant t = σ  ... 
doi:10.1080/10618600.2014.948181 pmid:27087770 pmcid:PMC4830509 fatcat:kkwuuro6yne3jbthgus5nlxo5e

Adaptive systems with closed-loop reference-models, part I: Transient performance

Travis E. Gibson, Anuradha M. Annaswamy, Eugene Lavretsky
2013 2013 American Control Conference  
Both Euclidean and L2 norms are used to characterize this transient behavior.  ...  This paper explores the transient properties of direct model reference adaptive control with closed loop reference models.  ...  constant associated with the exponential decay of e(t) which is derived from the upper bound on V from (20) and τ 2 is the time constant associated with A m in (2). a θ is a positive scalar that upper  ... 
doi:10.1109/acc.2013.6580353 fatcat:ex2nidav5bgf7dos7axj5ozhqi

Gradient Leakage Attack Resilient Deep Learning [article]

Wenqi Wei, Ling Liu
2021 arXiv   pre-print
which are closely aligned to the trend of gradient updates during differentially private model training.  ...  First, we analyze existing implementation of deep learning with differential privacy, which use fixed noise variance to injects constant noise to the gradients in all layers using fixed privacy parameters  ...  This is because the l2 updates across the T training iterations. norm of the layer-wise per-example gradient vector ∇W (t)im satisfies  ... 
arXiv:2112.13178v1 fatcat:yxjv3jlxnfd6tog5trjc3vfwxm

Automated Algorithmic Error Resilience for Structured Grid Problems Based on Outlier Detection

Amoghavarsha Suresh, John Sartori
2014 Proceedings of Annual IEEE/ACM International Symposium on Code Generation and Optimization - CGO '14  
into robust applications with the use of parser-based tools and minimal programmer effort.  ...  We demonstrate 2 × −3× improvement in output quality compared to the original algorithm with only 22% overhead, on average, for non-iterative structured grid problems.  ...  Unlike time-independent problems however, an appropriate number of grid updates for these problems can be decided without measuring a metric that tracks convergence (e.g., L2 norm error).  ... 
doi:10.1145/2581122.2544140 fatcat:6zfur6uspffffowslbumpomwbe

Continuous data assimilation for downscaling large-footprint soil moisture retrievals

Muhammad U. Altaf, Raghavendra B. Jana, Ibrahim Hoteit, Matthew F. McCabe, Christopher M. U. Neale, Antonino Maltese
2016 Remote Sensing for Agriculture, Ecosystems, and Hydrology XVIII  
A nudging term, estimated as the misfit between interpolants of the assimilated coarse grid measurements and the fine grid model solution, is added to the model equations to constrain the model's large  ...  This enables nudging of the model outputs towards values that honor the coarse resolution dynamics while still being generated at the fine scale.  ...  It is seen that at each step before converging with the black line, the model outputs have a norm slightly lower than the norm of the updated soil moisture fields from the previous time step.  ... 
doi:10.1117/12.2241042 fatcat:ekn7ijscmvf5heb4x6ibx3xnwy

Linearized Bregman iteration based model‐free adaptive sliding mode control for a class of non‐linear systems

Shouli Gao, Dongya Zhao, Xinggang Yan, Sarah K. Spurgeon
2020 IET Control Theory & Applications  
A Majorization-Minimization technique and the L 1 norm are used within the proposed optimization and an online iterative approach is described for update of the control law.  ...  The proposed methodology seeks to address issues with the computation of inverse matrices and problems with singularity in existing methods while at the same time seeking to enhance robustness.  ...  ACKNOWLEDGEMENTS This work is partially supported by the National Nature Science Foundation of China (61973315) and the Changjiang Scholars Program.  ... 
doi:10.1049/cth2.12039 fatcat:5shcgj5vtfcincezo4tqxioebi
« Previous Showing results 1 — 15 out of 6,820 results