Filters








15 Hits in 1.5 sec

S3Pool: Pooling with Stochastic Spatial Sampling [article]

Shuangfei Zhai, Hui Wu, Abhishek Kumar, Yu Cheng, Yongxi Lu, Zhongfei Zhang, Rogerio Feris
2016 arXiv   pre-print
We study this aspect and propose a novel pooling strategy with stochastic spatial sampling (S3Pool), where the regular downsampling is replaced by a more general stochastic version.  ...  We view the pooling operation in CNNs as a two-step procedure: first, a pooling window (e.g., 2× 2) slides over the feature map with stride one which leaves the spatial resolution intact, and second, downsampling  ...  We study this aspect and propose a novel pooling strategy with stochastic spatial sampling (S3Pool), where the regular downsampling is replaced by a more general stochastic version.  ... 
arXiv:1611.05138v1 fatcat:uinmplhbl5faxb74seyyesezeu

S3Pool: Pooling with Stochastic Spatial Sampling

Shuangfei Zhai, Hui Wu, Abhishek Kumar, Yu Cheng, Yongxi Lu, Zhongfei Zhang, Rogerio Feris
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
We study this aspect and propose a novel pooling strategy with stochastic spatial sampling (S3Pool), where the regular downsampling is replaced by a more general stochastic version.  ...  We view the pooling operation in CNNs as a twostep procedure: first, a pooling window (e.g., 2ˆ2) slides over the feature map with stride one which leaves the spatial resolution intact, and second, downsampling  ...  Pooling with Stochastic Spatial Sampling While the typical downsampling step of a max pooling layer intuitively reduces the spatial dimension of a feature map by always selecting the activations at fixed  ... 
doi:10.1109/cvpr.2017.426 dblp:conf/cvpr/ZhaiWKCLZF17 fatcat:socf2w6n25gr5agfwhbtwkjtdi

Implications of Pooling Strategies in Convolutional Neural Networks: A Deep Insight

Shallu Sharma, Rajesh Mehra
2019 Foundations of Computing and Decision Sciences  
It is believed that this work would help in extending the scope of understanding the significance of CNN along with pooling regimes for solving computer vision problems.  ...  This study presents a detailed review of the conventional and the latest strategies which would help in appraising the readers with the upsides and downsides of each strategy.  ...  [56] confirmed that S3Pool have ability to surpass even the dropout (with data augmentation) and the stochastic pooling approach with the marginal increment in the training time.  ... 
doi:10.2478/fcds-2019-0016 fatcat:6qif4l2rffhbtk54eyjvp263pq

Detail-Preserving Pooling in Deep Networks

Faraz Saeedan, Nicolas Weber, Michael Goesele, Stefan Roth
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition  
Inspired by the human visual system, which focuses on local spatial changes, we propose detailpreserving pooling (DPP), an adaptive pooling method that magnifies spatial changes and preserves important  ...  Importantly, its parameters can be learned jointly with the rest of the network.  ...  Specifically, we consider stochastic spatial sampling DPP (S3DPP), which extends S3Pool by first applying regular DPP with a stride of 1 to the input and then sampling rows and columns uniformly.  ... 
doi:10.1109/cvpr.2018.00949 dblp:conf/cvpr/SaeedanWG018 fatcat:jz2v36i7szcbzn3waxp2bpgili

Detail-Preserving Pooling in Deep Networks [article]

Faraz Saeedan, Nicolas Weber, Michael Goesele, Stefan Roth
2018 arXiv   pre-print
Inspired by the human visual system, which focuses on local spatial changes, we propose detail-preserving pooling (DPP), an adaptive pooling method that magnifies spatial changes and preserves important  ...  Importantly, its parameters can be learned jointly with the rest of the network.  ...  Specifically, we consider stochastic spatial sampling DPP (S3DPP), which extends S3Pool by first applying regular DPP with a stride of 1 to the input and then sampling rows and columns uniformly.  ... 
arXiv:1804.04076v1 fatcat:hncci467dvbmzpekxtwyeueaxu

Improving the Resolution of CNN Feature Maps Efficiently with Multisampling [article]

Shayan Sadigh, Pradeep Sen
2018 arXiv   pre-print
Methods such as stochastic pooling, fractional max-pooling, and S3Pool focus on regularizing CNNs by implicitly increasing the size of the dataset through stochastic pooling methods.  ...  Max-pooling and average-pooling is replaced with a stochastic pooling method that randomly samples an element from the pooling region according to a distribution given by the activities within the pooling  ... 
arXiv:1805.10766v1 fatcat:x7pwljhhqbb2rjxzklklslgqze

Discriminative Shape Feature Pooling in Deep Neural Networks

Gang Hu, Chahna Dixit, Guanqiu Qi
2022 Journal of Imaging  
Additionally, compared with existing models using either handcrafted features or deep network representations, our method not only improves the corresponding performance, but also reduces the computational  ...  While outperforming stochastic and S3Pool methods with notable margins, the DSF-Pooling-based models performed similarly to other approaches.  ...  Besides the conventional max and average pooling approaches, stochastic [34] and S3pool [35] pooling randomly select a node in a neighborhood, while other approaches either use multiple [36] and  ... 
doi:10.3390/jimaging8050118 fatcat:s5kxu7gsdzafpputkjefb6dn7u

Refining activation downsampling with SoftPool [article]

Alexandros Stergiou, Ronald Poppe, Grigorios Kalliatakis
2021 arXiv   pre-print
Experiments with pooling layer substitutions on ImageNet1K show an increase in accuracy over both original architectures and other pooling methods.  ...  An important feature of the pooling operation is the minimization of information loss, with respect to the initial activation maps, without a significant impact on the computation and memory overhead.  ...  (e) Stochastic Spatial Sampling (S3Pool) [46] samples random horizontal and vertical regions given a specified stride.  ... 
arXiv:2101.00440v3 fatcat:dbw4nrcjdjcvhkamgl7kzvlt6i

Hartley Spectral Pooling for Deep Learning [article]

Hao Zhang, Jianwei Ma
2018 arXiv   pre-print
Compared with FSP, the proposed spectral pooling avoids the use of complex arithmetic for frequency representation and reduces the computation.  ...  Such operation is commonly so-called pooling.  ...  Instead of picking the maximum values within each pooling region, stochastic pooling [17] and S3Pool [18] stochastically pick a node in each pooling region, and the former favors strong activations  ... 
arXiv:1810.04028v1 fatcat:g2ibcpoya5bs7n7sqyhsce7hdm

Stochastic Region Pooling: Make Attention More Expressive [article]

Mingnan Luo, Guihua Wen, Yang Hu, Dan Dai, Yingxue Xu
2019 arXiv   pre-print
In this work, we propose a novel method for channel-wise attention network, called Stochastic Region Pooling (SRP), which makes the channel descriptors more representative and diversity by encouraging  ...  Global Average Pooling (GAP) is used by default on the channel-wise attention mechanism to extract channel descriptors.  ...  For example, stochastic pooling [48] randomly selects the activation value based on a multinomial distribution formed by activations of each pooling region to regularize the network, and S3Pool [49]  ... 
arXiv:1904.09853v1 fatcat:6pfgr6zpo5afnkvpy5jjyw4qgu

AdaPool: Exponential Adaptive Pooling for Information-Retaining Downsampling [article]

Alexandros Stergiou, Ronald Poppe
2021 arXiv   pre-print
To this end, we propose an adaptive and exponentially weighted pooling method named adaPool.  ...  In contrast to common pooling methods, weights can be used to upsample a downsampled activation map. We term this method adaUnPool.  ...  (vi) Stochastic Spatial Sampling (S3Pool) [4] samples horizontal and vertical regions given a specified stride.  ... 
arXiv:2111.00772v2 fatcat:rpb22bomjbe5xb7k4vpj2djuxe

Hartley Spectral Pooling for Deep Learning

Hao Zhang & Jianwei Ma
2020 CSIAM Transactions on Applied Mathematics  
The proposed spectral pooling avoids the use of complex arithmetic for frequency representation, in comparison with Fourier pooling.  ...  Such operation is commonly called pooling.  ...  Instead of picking the maximum values within each pooling region, stochastic pooling [17] and S3Pool [18] stochastically pick a node in a window, and the former favors strong activations.  ... 
doi:10.4208/csiam-am.2020-0018 fatcat:4jdltgsk5bhrragwlulhhkm22q

Zernike Pooling: Generalizing Average Pooling Using Zernike Moments

Thomas Theodoridis, Kostas Loumponias, Nicholas Vretos, Petros Daras
2021 IEEE Access  
× 2, as well as the two variants of Stochastic pooling and AlphaMEX in every case.  ...  ., convolutional, normalization, regularization, pooling layers, etc.), with their main difference being the connectivity of these components within the architecture and not the components themselves.  ...  A value of α = 1 corresponds to average pooling, while α = 2 is equivalent to bilinear pooling. Stochastic spatial sampling (S3Pool), proposed by Zhai et al.  ... 
doi:10.1109/access.2021.3108630 fatcat:i2awdaxcibgt5jxhl47pnf365q

Regularized Pooling [article]

Takato Otsuzuki, Hideaki Hayashi, Yuchen Zheng, Seiichi Uchida
2020 arXiv   pre-print
with conventional pooling operations.  ...  In this paper, we propose regularized pooling, which enables the value selection direction in the pooling operation to be spatially smooth across adjacent kernels so as to compensate only for actual deformations  ...  [26] proposed S3Pool, which extends standard max pooling by decomposing pooling into two steps: max pooling with stride one and a non-deterministic spatial downsampling step by randomly sampling rows  ... 
arXiv:2005.03709v2 fatcat:fg266v4upvcotg7wlmlqcgvndy

Global learnable pooling with enhancing distinctive feature for image classification

Xingpeng Zhang, Xiaohong Zhang
2020 IEEE Access  
Besides, GLPool is not a hand-crafted pooling operation, which has the characteristic of adapting to any size of the input. With few parameters is added, GLPool is also a plug-and-play layer.  ...  Pooling layers appear widely in deep networks for its aggregating information in a local region and fast downsampling.  ...  In addition, various pyramid pooling methods are based on GAP, such as spatial pyramid pooling [28] , [34] , [35] , concentric circle pooling [29] . B.  ... 
doi:10.1109/access.2020.2997078 fatcat:opn3xdazmvcvxk7ijeo4gt43jq