Filters








59,630 Hits in 4.2 sec

High-Performance Long-Term Tracking with Meta-Updater [article]

Kenan Dai, Yunhua Zhang, Dong Wang, Jianhua Li, Huchuan Lu, and Xiaoyun Yang
2020 arXiv   pre-print
This work also introduces a long-term tracking framework consisting of an online local tracker, an online verifier, a SiamRPN-based re-detector, and our meta-updater.  ...  Long-term visual tracking has drawn increasing attention because it is much closer to practical applications than short-term tracking.  ...  In this work, we attempt to address this dilemma by designing a high-performance long-term tracker with a meta-updater.  ... 
arXiv:2004.00305v1 fatcat:w7om7weyavdwneqqljzk7kmdwy

High-Performance Long-Term Tracking With Meta-Updater

Kenan Dai, Yunhua Zhang, Dong Wang, Jianhua Li, Huchuan Lu, Xiaoyun Yang
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Long-term visual tracking has drawn increasing attention because it is much closer to practical applications than short-term tracking.  ...  Most top-ranked long-term trackers adopt the offline-trained Siamese architectures, thus, they cannot benefit from great progress of short-term trackers with online update.  ...  In this work, we attempt to address this dilemma by designing a high-performance long-term tracker with a meta-updater.  ... 
doi:10.1109/cvpr42600.2020.00633 dblp:conf/cvpr/DaiZWLLY20 fatcat:3vt7qzqdqvfvvaqpke6ilvv6pm

Visual Tracking by Adaptive Continual Meta-Learning

Janghoon Choi, Sungyong Baik, Myungsub Choi, Junseok Kwon, Kyoung Mu Lee
2022 IEEE Access  
INDEX TERMS Continual learning, meta learning, object tracking, visual tracking.  ...  In contrast to conventional meta-learning based approaches that regard visual tracking as an instance detection problem with a focus on finding good weights for model initialization, we consider both initialization  ...  where performance improvements are more pronounced in the long-term tracking applications.  ... 
doi:10.1109/access.2022.3143809 fatcat:ghc7qvhtafhohe26wzxblbbn7m

Learning to Update for Object Tracking with Recurrent Meta-learner

Bi Li, Wenxuan Xie, Wenjun Zeng, Wenyu Liu
2019 IEEE Transactions on Image Processing  
Equipped with our learned updater, the template-based tracker achieves state-of-the-art performance among realtime trackers on GPU.  ...  Model update lies at the heart of object tracking. Generally, model update is formulated as an online learning problem where a target model is learned over the online training set.  ...  The learning rate α is searched on OTB-2013 from 0.01 to 0.2 with step size 0.01. • SGD-based update: we adopt the short-term update and long-term update following MDNet 7 [40] .  ... 
doi:10.1109/tip.2019.2900577 fatcat:dmvrn6cwovbk3kx6wqhv36jvve

Online Meta-Learning for Scene-Diverse Waveform-Agile Radar Target Tracking [article]

Charles E. Thornton, R. Michael Buehrer, Anthony F. Martone
2021 arXiv   pre-print
We show that the meta-learning process results in an appreciably faster learning, resulting in significantly fewer lost tracks than a conventional learning approach equipped with an uninformative prior  ...  This paper studies a Bayesian meta-learning model for radar waveform selection which seeks to learn an inductive bias to quickly optimize tracking performance across a class of radar scenes.  ...  However, the performance benefits in terms of average performance are not as substantial as in terms of worst-case metrics, such as lost tracks.  ... 
arXiv:2110.11450v1 fatcat:2ucfonr4pje5vcyha4dheeal4u

Benchmarking Deep Trackers on Aerial Videos

Abu Md Niamul Taufique, Breton Minnehan, Andreas Savakis
2020 Sensors  
In our experiments, we use a subset of OTB2015 dataset with aerial style videos; the UAV123 dataset without synthetic sequences; the UAV20L dataset, which contains 20 long sequences; and DTB70 dataset  ...  In recent years, deep learning-based visual object trackers have achieved state-of-the-art performance on several visual object tracking benchmarks.  ...  For long-term tracking, an important consideration is consistent performance in a long temporal span, which tests the tracker's ability to create a robust model and perform efficient model updates.  ... 
doi:10.3390/s20020547 pmid:31963879 pmcid:PMC7014490 fatcat:3ern5fkpd5cydb5gj57wwnqvay

Recursive Least-Squares Estimator-Aided Online Learning for Visual Tracking [article]

Jin Gao, Yan Lu, Xiaojuan Qi, Yutong Kou, Bing Li, Liang Li, Shan Yu, Weiming Hu
2021 arXiv   pre-print
The recent few-shot online adaptation methods incorporate the prior knowledge from large amounts of annotated training data via complex meta-learning optimization in the offline phase.  ...  This helps the online deep trackers to achieve fast adaptation and reduce overfitting risk in tracking.  ...  So the aforementioned long-term LaSOT benchmark with high-quality dense annotations is proposed to bridge this gap, leading to the largest high-quality dense tracking benchmark which also provides a training  ... 
arXiv:2112.14016v1 fatcat:wphq2hwdgfdtvilmjisfncmebq

The impact of dissection and re-entry versus wire escalation techniques on long-term clinical outcomes in patients with chronic total occlusion lesions following percutaneous coronary intervention: An updated meta-analysis

Yejing Zhao, Hongyu Peng, Xiaonan Li, Jinghua Liu
2020 Cardiology Journal  
The meta-analysis was performed to evaluate the effect of dissection and re-entry (DR) vs. wire escalation (WE) techniques on long-term clinical outcomes in patients with chronic total occlusion (CTO)  ...  WE techniques, during the long-term follow-up.  ...  However, the long-term prognosis of patients with DR techniques remains controversial.  ... 
doi:10.5603/cj.a2020.0026 pmid:32104900 pmcid:PMC8169187 fatcat:qkettv4aubcklotq3cuk36zuei

Effect of early glycaemic control on HbA1c tracking and development of vascular complications after 5 years of childhood onset type 1 diabetes: Systematic Review and Meta‐analysis

Veena Mazarello Paes, Jessica K Barrett, David C Taylor‐Robinson, Heather Chesters, Dimitrios Charalampopoulos, David B Dunger, Russell M Viner, Terence J Stephenson
2019 Pediatric Diabetes  
A systematic review and meta-analysis was conducted to investigate if glycemic control measured by glycated hemoglobin (HbA1c) levels near diagnosis are predictive of future glycemic outcomes and vascular  ...  associated with long-term glycemic control.  ...  as glycemic control within 2 years of diagnosis of T1D) AND long-term glycemic tracking (defined as settling of HbA1c levels into long-term tracks of either > or <7% ie, 53 mmol/mol) and risk of future  ... 
doi:10.1111/pedi.12850 pmid:30932298 pmcid:PMC6701989 fatcat:nr6wg2m3dvbh3ddeng5ho7ty2e

Online Visual Tracking with One-Shot Context-Aware Domain Adaptation [article]

Hossein Kashiani, Amir Abbas Hamidi Imani, Shahriar Baradaran Shokouhi, Ahmad Ayatollahi
2021 arXiv   pre-print
The domain adaptation approach is backboned with only an off-the-shelf deep model.  ...  We further introduce a cost-sensitive loss alleviating the dominance of non-semantic background candidates over the semantic candidates, thereby dealing with the data imbalance issue.  ...  In the long-term strategy, updating is performed every τ int frames with the positive candidates gathered from the previous successful frames in the frame set M long .  ... 
arXiv:2008.09891v2 fatcat:2o57dm6rozggrin6pun6fwwtpq

DAWSON: A Domain Adaptive Few Shot Generation Framework [article]

Weixin Liang, Zixuan Liu, Can Liu
2020 arXiv   pre-print
DAWSON is a plug-and-play framework that supports a broad family of meta-learning algorithms and various GANs with architectural-variants.  ...  We also show that DAWSON can learn to generate new digits with only four samples in the MNIST dataset.  ...  The meta-performance is measured by evaluating the samples generated by the updated model in terms of their similarity with the target domain.  ... 
arXiv:2001.00576v1 fatcat:wjud4apyubdlzohgmiaw5qb4mm

An OAIS Based Approach to Effective Long-term Digital Metadata Curation

Arif Shaon, Andrew Woolf
2008 Computer and Information Science  
of long-term metadata curation in a comprehensive and unambiguous manner.  ...  Metadata has the proven ability to provide information necessary for successful long-term curation of digital objects.  ...  of that resource, along with accurate verification of its integrity (e.g. provenance tracking), its apposite (re-) use and effective preservation over the long-term.  ... 
doi:10.5539/cis.v1n2p2 fatcat:uqv5bnjmevfn7hr7vyfaf6rqse

Online Bayesian Meta-Learning for Cognitive Tracking Radar [article]

Charles E. Thornton, R. Michael Buehrer, Anthony F. Martone
2022 arXiv   pre-print
Finally, we examine the potential performance benefits and practical challenges associated with online meta-learning for waveform-agile tracking.  ...  One way to address this problem is to bias a learning algorithm strategically by exploiting high-level structure across tracking instances, referred to as meta-learning.  ...  This is because the radar begins each new track with updated prior assumptions, reducing data requirements. For this reason, meta-learning is intimately connected with Bayesian statistics.  ... 
arXiv:2207.06917v1 fatcat:jcpdvofn7zhmxlwea72teoxd3y

Updatable Siamese Tracker with Two-stage One-shot Learning [article]

Xinglong Sun, Guangliang Han, Lihong Guo, Tingfa Xu, Jianan Li, Peixun Liu
2021 arXiv   pre-print
Offline Siamese networks have achieved very promising tracking performance, especially in accuracy and efficiency.  ...  However, they often fail to track an object in complex scenes due to the incapacity in online update.  ...  For example, GradNet [23] presented a gradient-guided network to update the exemplar for Siamese network, while [5] proposed a meta-updater based on a memory network for long-term tracking.  ... 
arXiv:2104.15049v1 fatcat:wzhotgpqcfczlo6dqzlg6xbcye

PhilDB: the time series database with built-in change logging

Andrew MacDonald
2016 PeerJ Computer Science  
Recent open-source systems have been developed to indefinitely store long-period high-resolution time series data without change logging.  ...  PhilDB takes a unique approach to meta-data tracking; optional attribute attachment. This facilitates scaling the complexities of storing a wide variety of data.  ...  array data for high performance operations.  ... 
doi:10.7717/peerj-cs.52 fatcat:hgwp5zzp2vaoto6hqylndx4idm
« Previous Showing results 1 — 15 out of 59,630 results