arrow
返回

Robust visual tracking using very deep generative model

delete2023-01-11
delete6
delete
OA
AI
A
Abdel-Fattah Attia
A
Abdelmunim, Hossam E.
H
Hazem M. Abbas
DOI:10.1186/s40537-022-00682-4delete
delete原文链接
delete分享
delete收藏
摘要

摘要

En 中文
Deep learning algorithms provide visual tracking robustness at an unprecedented level, but realizing an acceptable performance is still challenging because of the natural continuous changes in the features of foreground and background objects over videos. One of the factors that most affects the robustness of tracking algorithms is the choice of network architecture parameters, especially the depth. A robust visual tracking model using a very deep generator (RTDG) was proposed in this study. We constructed our model on an ordinary convolutional neural network (CNN), which consists of feature extraction and binary classifier networks. We integrated a generative adversarial network (GAN) into the CNN to enhance the tracking results through an adversarial learning process performed during the training phase. We used the discriminator as a classifier and the generator as a store that produces unlabeled feature-level data with different appearances by applying masks to the extracted features. In this study, we investigated the role of increasing the number of fully connected (FC) layers in adversarial generative networks and their impact on robustness. We used a very deep FC network with 22 layers as a high-performance generator for the first time. This generator is used via adversarial learning to augment the positive samples to reduce the gap between the hungry deep learning algorithm and the available training data to achieve robust visual tracking. The experiments showed that the proposed framework performed well against state-of-the-art trackers on OTB-100, VOT2019, LaSOT and UAVDT benchmark datasets.
Keyword:
Deep learning
Generative adversarial network
Fully connected layers
Visual tracking

期刊

Journal of Big Data 封面图
Journal of Big Data
IF:
6.4
论文数:
1.5K
被引数:
1.4W

机构

E
egyptian knowledge bank (ekb)
学者数:
11.6W
论文数: 9.3W
被引数: 10.4W
K
kafr-elsheikh university
学者数:
7
论文数: 3
被引数: 0
引用论文

引用论文

Self-Supervised Deep Correlation Tracking
err2021-01-01
err222
PREAI
errYuan, Di; Chang, Xiaojun; Huang, Po-Yao; Liu, Qiao; He, Zhenyu
err分享
err收藏
Vision-based and marker-less surgical tool detection and tracking: a review of the literature
err2017-01-01
err217
errOAAI
errBouget, David; Allan, Max; Stoyanov, Danail; Jannin, Pierre
err分享
err收藏
Photothermal Therapy: Biodegradable Bi2O2Se Quantum Dots for Photoacoustic Imaging‐Guided Cancer Photothermal Therapy (Small 1/2020)
err2020-01-09
err0
errOAAI
errHanhan Xie; Mingqiang Liu; Baihao You; Guanghong Luo; Yue Chen; Bilu Liu; Zhenyou Jiang; Paul K. Chu; Jundong Shao; Xue‐Feng Yu
err分享
err收藏
Targeted mutation of zebrafish fga models human congenital afibrinogenemia
err2014-04-03
err0
errOAAI
errRichard J. Fish; Corinne Di Sanza; Marguerite Neerman-Arbez
err分享
err收藏
Dual-regression model for visual tracking
err2020-12-01
err24
PREAI
errLi, Xin; Liu, Qiao; Fan, Nana; Zhou, Zikun; He, Zhenyu; Jing, Xiao-yuan
err分享
err收藏
err分享
err收藏
Deep Learning for Visual Tracking: A Comprehensive Survey视觉跟踪的深度学习: 综合综述
err2022-05-01
err236
errOAAI
errMarvasti-Zadeh, Seyed Mojtaba; Cheng, Li; Ghanei-Yakhdan, Hossein; Kasaei, Shohreh
err分享
err收藏
学者 查看更多内容