Filters








68,770 Hits in 4.7 sec

Correlation Filters with Weighted Convolution Responses

Zhiqun He, Yingruo Fan, Junfei Zhuang, Yuan Dong, HongLiang Bai
2017 2017 IEEE International Conference on Computer Vision Workshops (ICCVW)  
Combining the online learning efficiency of the correlation filters with the discriminative power of CNN features has aroused great attention.  ...  In our work, we normalize each individual feature extracted from different layers of the deep pretrained CNN first, and after that, the weighted convolution responses from each feature block are summed  ...  Weights in Convolution Responses When the weights are assigned to the convolution responses produced by different convolutional layers, the spatial size and dimensionality of the convolutional features  ... 
doi:10.1109/iccvw.2017.233 dblp:conf/iccvw/HeFZDB17 fatcat:omlohrvbxffxjmcucjxhmuod4e

Deep Convolutional Likelihood Particle Filter for Visual Tracking [article]

Reza Jalil Mozhdehi, Henry Medeiros
2020 arXiv   pre-print
We propose a novel particle filter for convolutional-correlation visual trackers.  ...  Our method uses correlation response maps to estimate likelihood distributions and employs these likelihoods as proposal densities to sample particles.  ...  In these methods, particle filters sample several image patches and calculate the weight of each sample by applying a correlation filter to the convolutional response maps.  ... 
arXiv:2006.06746v1 fatcat:zseu5eozwbcfrags2lwe7c54ym

Deep Convolutional Particle Filter with Adaptive Correlation Maps for Visual Tracking

Reza Jalil Mozhdehi, Yevgeniy Reznichenko, Abubakar Siddique, Henry Medeiros
2018 2018 25th IEEE International Conference on Image Processing (ICIP)  
The robustness of the visual trackers based on the correlation maps generated from convolutional neural networks can be substantially improved if these maps are used to employed in conjunction with a particle  ...  In this article, we present a particle filter that estimates the target size as well as the target position and that utilizes a new adaptive correlation filter to account for potential errors in the model  ...  The likelihood or weight of each correlation response map is calculated by ( )( ) = ∑ =1 � ( , ) * ( )( ) =1 , (6) where ( , ) * ( )( ) refers to the element of the final correlation response map on row  ... 
doi:10.1109/icip.2018.8451069 dblp:conf/icip/MozhdehiRSM18 fatcat:5euv2jxu35fqvc37fczckspbni

Robust Visual Tracking via Hierarchical Convolutional Features [article]

Chao Ma, Jia-Bin Huang, Xiaokang Yang, Ming-Hsuan Yang
2018 arXiv   pre-print
Specifically, we learn adaptive correlation filters on the outputs from each convolutional layer to encode the target appearance.  ...  We infer the maximum response of each layer to locate targets in a coarse-to-fine manner.  ...  (b)-(d) Correlation response maps from single convolutional layers. (e)-(g) Different schemes to weight (b), (c) and (d).  ... 
arXiv:1707.03816v2 fatcat:rqxytlu64na4hmwtjcahk5otou

Robust Visual Tracking via Hierarchical Convolutional Features

Chao Ma, Jia-Bin Huang, Xiaokang Yang, Ming-Hsuan Yang
2018 IEEE Transactions on Pattern Analysis and Machine Intelligence  
Specifically, we learn adaptive correlation filters on the outputs from each convolutional layer to encode the target appearance.  ...  To further handle the issues with scale estimation and target re-detection from tracking failures caused by heavy occlusion or moving out of the view, we conservatively learn another correlation filter  ...  (b)-(d) Correlation response maps from single convolutional layers. (e)-(g) Different schemes to weight (b), (c) and (d).  ... 
doi:10.1109/tpami.2018.2865311 pmid:30106709 fatcat:dgpu2z5agrcc5mtxtmhvjaxtgu

Convolutional neural network based natural image and MRI classification using Gaussian activated parametric (GAP) layer

Bijen Khagi, Goo Rak Kwon
2021 IEEE Access  
Xf is now spatially correlated with a Gaussian kernel, i.e., filtering is done with the kernel of a size equivalent to the size of the previous convolution filter (here 3×3×3) for the first GAP layer as  ...  In CNN, the weights of convolution filters are the key parameter to train, and it determines how a particular filter works.  ...  The prepared dataset can be downloaded from Author Name: Preparation of Papers for IEEE Access (February 2017) VOLUME XX, 2017 Appendix C1: MATLAB code implementation for GAP layer used after convolutional  ... 
doi:10.1109/access.2021.3093455 fatcat:tr2idmi7hzdxtmch2j3y3uckb4

Reconstructed spatial receptive field structures by reverse correlation technique explains the visual feature selectivity of units in deep convolutional neural networks [article]

Yoshiyuki R Shiraishi, Hiromichi Sato, Takahisa M Sanada, Tomoyuki Naito
2021 arXiv   pre-print
The spatial structures of the receptive fields of all convolutional units were estimated by activation-weighted average (AWA) and activation-weighted covariance (AWC) analyses.  ...  An important issue in dealing with Deep Convolutional Neural Networks (DCNN) is the 'black box problem', which represents the unknowns about internal information representation and processing, especially  ...  We also predicted the measured response with only an AWA filter and the LN model.  ... 
arXiv:2103.02587v1 fatcat:ryenloikh5fspkboqbu6p5kbqu

CREST: Convolutional Residual Learning for Visual Tracking

Yibing Song, Chao Ma, Lijun Gong, Jiawei Zhang, Rynson W.H. Lau, Ming-Hsuan Yang
2017 2017 IEEE International Conference on Computer Vision (ICCV)  
However, existing DCFs learn the filters separately from feature extraction, and update these filters using a moving average operation with an empirical weight.  ...  Discriminative correlation filters (DCFs) have been shown to perform superiorly in visual tracking.  ...  The spatial convolutional operation functions similarly with the dot product between the circulant shifted inputs and the correlation filter.  ... 
doi:10.1109/iccv.2017.279 dblp:conf/iccv/SongMGZL017 fatcat:f5v3wttxenhanl56b6dhgzijgm

CREST: Convolutional Residual Learning for Visual Tracking [article]

Yibing Song, Chao Ma, Lijun Gong, Jiawei Zhang, Rynson Lau, Ming-Hsuan Yang
2017 arXiv   pre-print
However, existing DCFs learn the filters separately from feature extraction, and update these filters using a moving average operation with an empirical weight.  ...  Discriminative correlation filters (DCFs) have been shown to perform superiorly in visual tracking.  ...  The spatial convolutional operation functions similarly with the dot product between the circulant shifted inputs and the correlation filter.  ... 
arXiv:1708.00225v1 fatcat:425ey5aq6vbzbiu6l254vvgeya

Robust visual tracking via multilayer CaffeNet features and improved correlation filtering

Yuqi Xiao, Difu Pan
2019 IEEE Access  
INDEX TERMS Convolutional neutral network, correlation filter, target tracking, computer vision technology.  ...  For problems related to the robust tracking of visual objects in various challenging tracking conditions, a robust visual tracking method based on multilayer convolutional features and correlation filtering  ...  This means that the zeroing probability of convolution kernels with larger weights should be less than that of convolution kernels with smaller weights.  ... 
doi:10.1109/access.2019.2957518 fatcat:rfu4mczfo5cxdfxchh6wwqchrq

Feature Adaptive Correlation Tracking

Yulong XU, Yang LI, Jiabao WANG, Zhuang MIAO, Hang LI, Yafei ZHANG
2017 IEICE transactions on information and systems  
Furthermore, we employ a discriminative correlation filter to handle scale variations. Extensive experiments are performed on a large-scale benchmark challenging dataset.  ...  According to the luminance of the target, our approach automatically selects either hierarchical convolutional features or histogram of oriented gradient features in translation for varied scenarios.  ...  Therefore, the final correlation response map ŷ is the linear combination of the three correlation response maps, ŷ = 3 l=1 γ l ŷl , (3) where γ l is the weight of the different response map ŷl .  ... 
doi:10.1587/transinf.2016edl8164 fatcat:2kpn7f4x35d5pe7y4bgoba64t4

Learning Local–Global Multiple Correlation Filters for Robust Visual Tracking with Kalman Filter Redetection

Jianming Zhang, Yang Liu, Hehua Liu, Jin Wang
2021 Sensors  
First, we construct a global correlation filter model with deep convolutional features, and choose horizontal or vertical division according to the aspect ratio to build two local filters with hand-crafted  ...  Multilayer convolutional features comprehensively used in correlation filter (CF)-based tracking algorithms have achieved excellent performance.  ...  With the widespread application of convolutional neural networks [28, 29] , the hierarchical convolutional features tracker (HCF) [17] and Correlation Filters with Weighted Convolution Responses (CFWCR  ... 
doi:10.3390/s21041129 pmid:33562878 fatcat:2yxnngieu5huxjmpaep4vp2sd4

Improved C-COT based on feature channels confidence for visual tracking

Shuwei SHEN, Shengjing TIAN, Leyuan WANG, Aihong SHEN, Xiuping LIU
2019 Journal of Advanced Mechanical Design, Systems, and Manufacturing  
The Continuous Convolution Operator Tracker (C-COT) is a novel correlation filter to track the target position in the continuous domain, which achieved significant effects.  ...  The Average Peak Correlation Energy (APCE) is used to evaluate the corresponding response map of each feature channel, guiding the target appearance model to give different weights to different features  ...  Firstly, The average peak correlation energy is used to evaluate the response map corresponding to each feature block, and this guides the appearance model to give different weights to different filters  ... 
doi:10.1299/jamdsm.2019jamdsm0096 fatcat:sbbw2dsgsbbabbvalczhz34kaq

Visual Tracking via Deep Feature Fusion and Correlation Filters

Haoran Xia, Yuanping Zhang, Ming Yang, and Yufang Zhao
2020 Sensors  
This paper builds a hybrid tracker combining the deep feature method and correlation filter to solve this challenge, and verifies its powerful characteristics.  ...  Specifically, an effective visual tracking method is proposed to address the problem of low tracking accuracy due to the limitations of traditional artificial feature models, then rich hiearchical features of Convolutional  ...  The feature output of each convolutional layer of CNNs is then taken as the target object, and the convolution operation is performed again with the learned correlation filter to generate the response  ... 
doi:10.3390/s20123370 pmid:32545916 pmcid:PMC7349342 fatcat:wzt4qnuxjbgtflwdgkn5ki4md4

Dual Model Learning Combined with Multiple Feature Selection for Accurate Visual Tracking

Jianming Zhang, Xiaokang Jin, Juan Sun, Jin Wang, Keqin Li
2019 IEEE Access  
Over these years, object tracking algorithms combined with correlation filters and convolutional features have achieved excellent performance in accuracy and real-time speed.  ...  First, we fuse the handcrafted features with the multi-layer features extracted from the convolutional neural network to construct a correlation filter learning model, which can precisely localize the  ...  We also construct the response maps from the low-level features after obtaining responses of convolutional layer, thus forming correlation response maps with six layers (including five convolutional layer  ... 
doi:10.1109/access.2019.2908668 fatcat:3k3y3zsu4zda7ajjwf2dfx25z4
« Previous Showing results 1 — 15 out of 68,770 results