Filters








2,592 Hits in 4.7 sec

Efficient Scale Estimation Methods using Lightweight Deep Convolutional Neural Networks for Visual Tracking [article]

Seyed Mojtaba Marvasti-Zadeh, Hossein Ghanei-Yakhdan, Shohreh Kasaei
2020 arXiv   pre-print
Although a wide range of recent DCF-based methods exploit the features that are extracted from deep convolutional neural networks (CNNs) in their translation model, the scale of the visual target is still  ...  Whereas the exploitation of CNNs imposes a high computational burden, this paper exploits pre-trained lightweight CNNs models to propose two efficient scale estimation methods, which not only improve the  ...  To address the limitation of using deep neural networks for robust scale estimation, twoscale estimation methods using lightweight CNNs are proposed.  ... 
arXiv:2004.02933v2 fatcat:p3sqisjjpvhbrcyzceknelefnm

VCIP 2020 Index

2020 2020 IEEE International Conference on Visual Communications and Image Processing (VCIP)  
Deep Convolutional Neural Network Based on Multi-Scale Feature Extraction for Image Denoising Zhang, Jingyi Machine Learning for Photometric Redshift Estimation of Quasars with Different Samples  ...  T S Sang, Liu Deep Convolutional Neural Network Based on Multi-Scale Feature Extraction for Image Denoising Santos, João M.  ... 
doi:10.1109/vcip49819.2020.9301896 fatcat:bdh7cuvstzgrbaztnahjdp5s5y

RPEOD: A Real-Time Pose Estimation and Object Detection System for Aerial Robot Target Tracking

Chi Zhang, Zhong Yang, Luwei Liao, Yulong You, Yaoyu Sui, Tang Zhu
2022 Machines  
In this paper, a novel real-time pose estimation and object detection (RPEOD) strategy for aerial robot target tracking is presented.  ...  The aerial robot is equipped with a binocular fisheye camera for pose estimation and a depth camera to capture the spatial position of the tracked target.  ...  Acknowledgments: The authors would like to acknowledge Qiuyan Zhang for his great support and reviews.  ... 
doi:10.3390/machines10030181 fatcat:i2uv3tkwfvaezowf6m7hv4gqmq

SiamMixer: A Lightweight and Hardware-Friendly Visual Object-Tracking Network

Li Cheng, Xuemin Zheng, Mingxin Zhao, Runjiang Dou, Shuangming Yu, Nanjian Wu, Liyuan Liu
2022 Sensors  
In this paper, we present SiamMixer, a lightweight and hardware-friendly visual object-tracking network.  ...  Benefiting from these techniques, SiamMixer demonstrates a comparable accuracy to other large trackers with only 286 kB parameters and 196 kB extra memory use for feature maps.  ...  Designing efficient and lightweight neural networks for mobile devices has attracted much attention in the past few years.  ... 
doi:10.3390/s22041585 pmid:35214487 pmcid:PMC8876980 fatcat:5i7tgot4hvaypon4mfejfwblju

2021 Index IEEE Transactions on Pattern Analysis and Machine Intelligence Vol. 43

2022 IEEE Transactions on Pattern Analysis and Machine Intelligence  
The Author Index contains the primary entry for each item, listed under the first author's name.  ...  Zhou, Y., +, TPAMI Jan. 2021 62-76 BlockQNN: Efficient Block-Wise Neural Network Architecture Generation. Zhong, Z., +, TPAMI July 2021 2314-2328 Deep Affinity Network for Multiple Object Tracking.  ...  ., +, TPAMI May 2021 1777-1790 Computational efficiency ChannelNets: Compact and Efficient Convolutional Neural Networks via Channel-Wise Convolutions.  ... 
doi:10.1109/tpami.2021.3126216 fatcat:h6bdbf2tdngefjgj76cudpoyia

Visual tracking based on transfer learning of deep salience information

Zuo Haorui, Xu Zhiyong, Zhang Jianlin, Jia Ge
2020 Opto-Electronic Advances  
In this paper, we propose a new visual tracking method in light of salience information and deep learning. Salience detection is used to exploit features with salient information of the image.  ...  Complicated representations of image features can be gained by the function of every layer in convolution neural network (CNN).  ...  We express our thanks for the experiment equipment provided by the lab. We appreciate the support of the relevant department.  ... 
doi:10.29026/oea.2020.190018 doaj:e230fac26e2948d28ca973523f087436 fatcat:skxrt4q2yrdeheanqjpwfyjnau

Lightweight Cross-fusion Network on Human Pose Estimation for Edge Device

Xian Zhu, Xiaoqin Zeng, Wei Ma
2021 IEEE Access  
Due to memory and storage space limitations, it is difficult for edge devices to maintain implementing Convolutional Neural Networks, which deployed large-scale terminal platforms with abundant computing  ...  Using state-of-the-art efficient neural architecture, and Ghost Net, as the backbone, which are gradually applying a cross-information fusion network for key points extraction in the baseline and strengthen  ...  Our method showed that the network can run on edge devices, and Ghost Net is a powerful and efficient backbone for human pose estimation.  ... 
doi:10.1109/access.2021.3065574 fatcat:r2zl54crdnb5dgtzqdm6owoxzi

2020 Index IEEE Transactions on Image Processing Vol. 29

2020 IEEE Transactions on Image Processing  
., +, TIP 2020 5900-5911 Efficient In-Loop Filtering Based on Enhanced Deep Convolutional Neural Networks for HEVC.  ...  ., +, TIP 2020 9678-9688 Efficient In-Loop Filtering Based on Enhanced Deep Convolutional Neural Networks for HEVC.  ... 
doi:10.1109/tip.2020.3046056 fatcat:24m6k2elprf2nfmucbjzhvzk3m

2021 Index IEEE Transactions on Image Processing Vol. 30

2021 IEEE Transactions on Image Processing  
The Author Index contains the primary entry for each item, listed under the first author's name.  ...  ., +, TIP 2021 3204-3216 Resolution Learning in Deep Convolutional Networks Using Scale-Space Theory.  ...  Chu, R.J., +, TIP 2021 4341-4356 Resolution Learning in Deep Convolutional Networks Using Scale-Space Theory.  ... 
doi:10.1109/tip.2022.3142569 fatcat:z26yhwuecbgrnb2czhwjlf73qu

Front Matter: Volume 10396

Andrew G. Tescher
2017 Applications of Digital Image Processing XL  
using a Base 36 numbering system employing both numerals and letters.  ...  Publication of record for individual papers is online in the SPIE Digital Library. SPIEDigitalLibrary.org Paper Numbering: Proceedings of SPIE follow an e-First publication model.  ...  deep convolutional neural networks and end-to-end learning [10396-74] 10396 23 A locally adaptive algorithm for shadow correction in color images [10396-75] 10396 24 Tracking of multiple objects  ... 
doi:10.1117/12.2293188 fatcat:4uko25br6rcmzjvi4ti2abf4re

Deep Learning based Pedestrian Inertial Navigation: Methods, Dataset and On-Device Inference [article]

Changhao Chen, Peijun Zhao, Chris Xiaoxuan Lu, Wei Wang, Andrew Markham, Niki Trigoni
2020 arXiv   pre-print
Recently, there has been a growing interest in applying deep neural networks (DNNs) to motion sensing and location estimation.  ...  Furthermore, to enable more efficient inference at the edge, we propose a novel lightweight framework to learn and reconstruct pedestrian trajectories from raw IMU data.  ...  Lightweight Inertial Odometry Neural Networks To this end, we introduce the Lightweight Inertial Odometry Neural Network (L-IONet), a lightweight framework to learn inertial tracking, which is more efficient  ... 
arXiv:2001.04061v1 fatcat:6vzynfiklza7la4dnejyodoyky

Siamese Attentional Keypoint Network for High Performance Visual Tracking [article]

Peng Gao, Yipeng Ma, Ruyue Yuan, Liyi Xiao, Fei Wang
2019 arXiv   pre-print
Firstly, a new Siamese lightweight hourglass network is specifically designed for visual tracking.  ...  Network, dubbed SATIN, to achieve efficient tracking and accurate localization.  ...  a few approaches take full use of these relationships efficiently in visual tracking.  ... 
arXiv:1904.10128v1 fatcat:ovn4s5cbdfbkllioy7fxnqxp5q

Adoption of Convolutional Neural Network Algorithm Combined with Augmented Reality in Building Data Visualization and Intelligent Detection

Minghui Wei, Jingjing Tang, Haotian Tang, Rui Zhao, Xiaohui Gai, Renying Lin, Zhihan Lv
2021 Complexity  
Convolutional neural network and augmented reality technology are adopted, and a building visualization model based on convolutional neural network and augmented reality is proposed.  ...  It is found that the building target detection model based on convolutional neural network and augmented reality has obvious advantages in algorithm complexity and recognition accuracy.  ...  target detection model SqueezeNet SSD is designed for mobile. e model is based on the deep learning model SSD, combined with the SqueezeNet lightweight convolutional neural classification network to reduce  ... 
doi:10.1155/2021/5161111 fatcat:6lzdhwxtercvzbh2qpdypear3q

Front Matter: Volume 11373

Zhigeng Pan, Xun Wang
2020 Eleventh International Conference on Graphics and Image Processing (ICGIP 2019)  
using a Base 36 numbering system employing both numerals and letters.  ...  Publication of record for individual papers is online in the SPIE Digital Library. SPIEDigitalLibrary.org Paper Numbering: Proceedings of SPIE follow an e-First publication model.  ...  of extracting target trajectory by deep convolution network in infrared images [11373-104] 11373 11 A deep convolutional neural network-based low-light image enhancement using illumination map [11373  ... 
doi:10.1117/12.2561685 fatcat:5dwsd2oxjjcdllprs6rtum4tfu

Convolutional Neural Networks(CNN) based Eye-Gaze Tracking System using Machine Learning Algorithm

Prakash Kanade, Fortune David, Sunay Kanade
2021 European Journal of Electrical Engineering and Computer Science  
A deep learning-based gaze estimation technique has been considered to solve this issue, with an emphasis on WSN based Convolutional Neural Networks (CNN) based system.  ...  This work also includes pre-trained models, network structures, and datasets for designing and developing CNN-based deep learning models for Eye-Gaze Tracking and Classification.  ...  A deep learning-based gaze estimation technique has been considered to solve this challenge, with an emphasis on Convolutional Neural Networks (CNN) based methods.  ... 
doi:10.24018/ejece.2021.5.2.314 fatcat:u2khdsussrdhtfclliqclw56xy
« Previous Showing results 1 — 15 out of 2,592 results